audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
Our simple 500W inverter project has a square-wave output that has the same effective power as a sine-wave.
http://www.electronics-lab.com/projects/power/033/index.html
It uses an IC to make two 50Hz signals, one is the opposite phase of the other. Then the circuit uses a couple of opamps and a bunch of transistors to amplify the current from the oscillator up to 50A. The transistors drive a 12V-0-12V transformer backwards so it increases the voltage up to mains level. If available, 2 or 4 Mosfets can replace all those transistors.
http://www.electronics-lab.com/projects/power/033/index.html
It uses an IC to make two 50Hz signals, one is the opposite phase of the other. Then the circuit uses a couple of opamps and a bunch of transistors to amplify the current from the oscillator up to 50A. The transistors drive a 12V-0-12V transformer backwards so it increases the voltage up to mains level. If available, 2 or 4 Mosfets can replace all those transistors.
i.e. 4047 produces 2 square wave ACs and you are amplifying it by op-amps and further by set of transistors. and finally feeding it to a 12-0-12 to 220-0 transformer.
My question : Is there any IC which produces sinewave signals (2 sets opposite in phase to each other)?
What is the difference in using this type of inverter compared with the sine-one?
What if I use 555 and amplify using the same topology and finally feeding it to a 12-0 to 220-0 type transformer?
My question : Is there any IC which produces sinewave signals (2 sets opposite in phase to each other)?
What is the difference in using this type of inverter compared with the sine-one?
What if I use 555 and amplify using the same topology and finally feeding it to a 12-0 to 220-0 type transformer?
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
No, but the transistors would melt if they were fed a sine-wave anyway. The transistors switch fully on (low voltage so low heating) and fully off (no current so no heating) with a square-wave. A linear sine-wave would keep them conducting near halfway all the time and make them extremely hot. Also, a linear sine-wave inverter would use more battery current than a square-wave inverter.vivek_pv said:
A sine-wave can be made with high frequency Pulse-Width-Modulation to make an efficient sine-wave inverter.
Some motor speed controls don't work with a square-wave. Most electronic products receive a voltage that is too low. Multi-voltage radios and TVs, and lights and heaters are fine with a square-wave.
A 555 uses more parts and costs much more than a CD4047. The signal to the transformer would be pulsing DC instead of AC causing the transformer to saturate if it is loaded. The transistors operate only half the time. The output power would be about 1/4 or less the power rating of the transformer.
Ok...
For a general purpose lighting inverter, is the model with IC NE555 be useful?
The technique will be:
IC 555--->Square Wave AC---->Amplify------>Transformer---->output.
Will it work this way?
Which one do you prefer, 12-0-12 to 220-0 (backwards) or 0-12 to 220-0 (backwards) for this case?
For a general purpose lighting inverter, is the model with IC NE555 be useful?
The technique will be:
IC 555--->Square Wave AC---->Amplify------>Transformer---->output.
Will it work this way?
Which one do you prefer, 12-0-12 to 220-0 (backwards) or 0-12 to 220-0 (backwards) for this case?
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
The 555 doesn't make AC, it makes pulsing DC. A coupling capacitor can feed an ordinary transformer winding from a complementary push-pull output circuit.vivek_pv said:
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
A 555 astable oscillator makes nearly a square-wave. With more resistors and a diode then the output can be adjusted to be a perfect square-wave.vivek_pv said:
The 4047 needs only a single resistor and a single capacitor to make a perfect square-wave and its opposite phase.
There are lots of circuits that make nearly a square-wave but they use lots of parts.
Many designs use Buffers, many use Inverters (NOT gates) to produce a nice square wave. I have also seen AND and NAND gates used to do this. Digital gates are good for providing a square wave. In fact, many circuit designs employ them to square up other wave forms that are not clean enough for a good digital switching action. The nice thing about using digital gates is that they have a good fan-out. Thus, you can parallel and/or series them to get a stronger signal if ever needed.vivek_pv said:
MP
Last edited by a moderator:
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
Aren't we talking about an power inverter circuit, not just any oscillator?MP said:
Their waveform isn't a perfect square-wave. It is "squared" but is not 50-50 duty-cycle.
A 4047 oscillator makes a "squared-wave" signal like those others then divides it by 2 with a flip-flop so its outputs are exactly a square-wave, with a perfect 50-50 duty cycle. With just a single resistor and a single capacitor to set the frequency.
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
It is an ordinary Cmos logic IC. Their supply voltage range is from 3V to 18V and their output voltage goes to the supply voltage if the load current is very low. With a 13.8V supply, the max output current is about 8mA.vivek_pv said:
A UPS charges its batteries when power is present then automatically switches to power the load when the power fails. Then it switches back again when the power is restored.
Ok...
In the cited design of an inverter, you have used LM358 and then the output is fed to one darlington pair and subsequent transistors. Instead, if I use a series of op-amps like LM358s, then does it work that way?
If I want to use the MOSFETs, then what are suitable substitutes?
In the cited design of an inverter, you have used LM358 and then the output is fed to one darlington pair and subsequent transistors. Instead, if I use a series of op-amps like LM358s, then does it work that way?
If I want to use the MOSFETs, then what are suitable substitutes?
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
A series of opamps won't do anything. Their max output current is only 20mA.vivek_pv said:
Mosfets switch very quickly. The CD4047 or the opamp switch slower and might cause both Mosfet sides to conduct at the same time. Investigation is required.
Many things like current and heat dissipation specs and spike voltage protection must be considered when designing an inverter that uses Mosfets.
Mosfets available in my country might not be available in your country.
There is nothing in the datasheet that proclaims a "perfect 50-50 duty cycle. In fact, the manufacturer puts a disclaimer in the data sheet that a 50% duty cycle is not guaranteed. Fortunately, for a square wave inverter, you do not need a perfect square wave. You don't even need a perfect 50 hz or 60 hz. The square wave inverters are the worst of all inverters out there. You will see a large variance from one to the other. If you want to build something that will be useful, build a sinewave inverter.audioguru said:
MP
View attachment 39866
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
The CD4047 has 3 outputs:
1) Its oscillator output which is an ordinary Cmos oscillator and they say "a 50% duty-cycle is not guaranteed". Inverter circuits don't use it.
2) a Q output at half the oscillator frequency and they say "at 50% duty cycle".
3) A Q-not output at half the oscillator frequency and they say "at 50% duty cycle".
A flip-flop divides the oscillator so the Q and Q-not outputs have perfect 50-50 duty cycle.
1) Its oscillator output which is an ordinary Cmos oscillator and they say "a 50% duty-cycle is not guaranteed". Inverter circuits don't use it.
2) a Q output at half the oscillator frequency and they say "at 50% duty cycle".
3) A Q-not output at half the oscillator frequency and they say "at 50% duty cycle".
A flip-flop divides the oscillator so the Q and Q-not outputs have perfect 50-50 duty cycle.
audioguru2
- Apr 6, 2004
- 12,026
- Joined
- Apr 6, 2004
- Messages
- 12,026
The datasheet says clearly exactly what I said. I even explained how the IC uses a digital divider to make two of its outputs perfect square-waves.MP said: