Are there any way of turning DC to AC

[jacquesl]

Are there any way of turning DC to AC, without the use of transistors or relay’s
Only caps and transistors?

[Cobra2]

oscillator ?

Switch-mode psu?

Arne K

[pinkmouse]

I assume you mean caps and resistors. Electronically, no. You could do it mechanically though.

Jacques, please buy a couple of basic physics and electronics textbooks.

[jacquesl]

Originally posted by jacquesl

Only caps and transistors?

Lol, sorry, it’s actually supposed to be “Caps and Transformers” and No tunnel diodes, It looks like one of those pizza eating and typing moments.

Yip I did one mechanically it an oscillator with a relay and a cap it and some wiring. Is variable then changing the voltage or cap size, but it’s very unstable and sensitive, and the relay can cause an interference on the low frequency band

But can this be done with a small 1:1 transformer.

[zigzagflux]

99% sure without using any type of switching device (electronic or mechanical) you are limited to a generator.

If you get a sort-of good AC from some other method, putting this through a ferroresonant transformer ($$$$$) will go a long way to clean it up into a fairly good sine wave.

You may find frequency regulation to be your biggest problem without active switches.

May we all ask, WHY?

[jacquesl]

It doesn’t have to be a sine wave, I’ll choose square wave to begin with.

I basically want to build an inverter that uses the minimal amount of parts and also the cheapest and it must give a usable clean AC output.

And to get the 12V to 230V, there’s two ways.

(1) Will be to use a transformer
(2) Will be to use caps in parallel charge and series dumping

What will be the cheapest?

And also the stable section, to keep the voltage fixed at a given value, like 230VAC or 110VAC, how is this done?

[Electrone]

The simplest solid state way is similar to the old technique of using two transistors driving a transformer at 60hz. In the old days the circuit self-oscillated and used bipolar power transistors. MOSFETs would be an improvement in the methodology. You could use a 555 timer, but I would use a CD40106 to produce a square wave close to 60hz. Run the output of the oscillator gate of the CD40106 to another CD40106 gate to buffer it to drive one MOSFET gate, and a pair of CD40106 gates in series to drive the other MOSFET gate. Both MOSFETs drive a center tap transformer. The center tap connects to 12VDC.

You would need just the right transformer for the task. You could probably take an old microwave oven transformer, cut off the high voltage secondary winding, and then wind your own center tap winding in such a way to obtain the needed voltage on the original primary. If you need +/- output voltages, you can use a diode/capacitor voltage doubler. Keep studying your basics, too.

Someone on my group also did an inverter circuit. The Internet host of the group does not allow non-members to view the files so it would not help much for me to post a link to diagrams.

[jacquesl]

Yip, I have a variable 555 IC oscillator, I just it for some of my experiments
I’ve seen that you can use something like 17Khz and just 6 windings and it’s just as good as thousand of turns at 60hz, do you know a formula to calculate the frequency and windings ratio’s, I’ve also noticed that the transistor heats up very quickly, I will like to know why? I use the 2N3055 ones and “I use a heat sink”, if anyone asks

[Electrone]

Yip, I have a variable 555 IC oscillator, I just it for some of my experiments
I’ve seen that you can use something like 17Khz and just 6 windings and it’s just as good as thousand of turns at 60hz, do you know a formula to calculate the frequency and windings ratio’s, I’ve also noticed that the transistor heats up very quickly, I will like to know why? I use the 2N3055 ones and “I use a heat sink”, if anyone asks

It should work much better for you to use MOSFETs instead of bipolar transistors. I have determined minimum primary turns needed by adding some turns to the transformer core. Use a current limiting power supply and then drive the core with a bridge circuit. If the no-load impedance is too low, then add some turns. Give it some extra margin. You don't want to drive the transformer at the point of no-load saturation as I figure things. I could be wrong here, but I have done it this way. Most of my circuits have plenty of primary inductance, so it never becomes an issue for me.

Full bridge at 12v is not efficient like push-pull, but for lazy work to get quick results, it is easier than dealing with flyback spikes from leakage inductance.

I have seen a 12v power supply using what has been commonly known as "flyback" used in old Zenith television sets, but they have trouble with low efficiency.

[KISS]

I did two designs about 30 years ago based on a 555 and a couple of 2N3055's. The 555 was set as a 50% duty cycle oscillator. A small signal transistor created the inversion for the other side.

1. A solid state vibrator for a Blaupunkt tube car radio.
2. The other was for an inverter for a telescope. Frequency was important.