High Speed, Unipolar Sine Inverter?

Thread Starter

johnyradio

Joined Oct 26, 2012
434
OK, now the picture starts to come into focus ..........
This sounds like a Homework assignment.
.
.
.
No, it's not a homework assignment. I'm 58 years old and not in school. My nonprofit org is developing a project. www.theboom.org
Note, i posted that "I'm fine doing this with all off-the-shelf modules." Obviously, that wouldn't be acceptable for a homework assignment.
Seriously, you cannot make assumptions like that. Feel free not to participate in the thread if you're paranoid.
 
Last edited:

LowQCab

Joined Nov 6, 2012
4,023
Then answer all the questions in post #13 and You will start getting valuable responses.
If You don't present the entire picture, then it's all rather pointless.
.
.
.
 

LowQCab

Joined Nov 6, 2012
4,023
The picture of a Sine-Wave shows a 1-Volt Peak to Peak waveform, with a 0.5-Volt Positive-Offset.
This makes the Output being specified a varying DC-Voltage.

And, your post says nothing about the characterization of the Load.
But it does say that You want 20-Amps, instead of the 5-Amps originally specified.

If You can't properly characterize the Load,
or at least give the best description of the Load that You can,
then there's no telling what the "guesses" might be that You will get.

25-Watts, as opposed to 100-Watts, is a huge jump, and changes everything.
.
.
.
 

ronsimpson

Joined Oct 7, 2019
2,987
Our coupled inductor is for galvanic isolation between this PS and the load.
Inductor for isolation...... how? If the power supply has isolation will that help?

Normally when you make a waveform and measure it, 0V in the center and a 5Vrms waveform has +/-7V peak.
Do you want to have a peak at 0V and the other peak to be at 14V? (see below)

How regulated must the voltage be? Will +/-20% work or it can be 1%.
Do you need current limit for when the load is shorted?
How accurate must the "0V" be? Can it be -0.6V?
1679354736182.png
Sorry but we do not understand. I design circuits for a living. It happens that people want something but will not tell all the details. Secret. There is a good chance any circuit will not work the way you want.
 

Thread Starter

johnyradio

Joined Oct 26, 2012
434
The picture of a Sine-Wave shows a 1-Volt Peak to Peak waveform, with a 0.5-Volt Positive-Offset.
This makes the Output being specified a varying DC-Voltage.

your post says nothing about the characterization of the Load.
But it does say that You want 20-Amps, instead of the 5-Amps originally specified.
I updated the original question. And for simplicity i'll stick with the original 5A.
 
Last edited:

Thread Starter

johnyradio

Joined Oct 26, 2012
434
Do you want to have a peak at 0V and the other peak to be at 14V? (see below)
How regulated must the voltage be? Will +/-20% work or it can be 1%.
Do you need current limit for when the load is shorted?
How accurate must the "0V" be? Can it be -0.6V?
Answered in the original question.
 

Thread Starter

johnyradio

Joined Oct 26, 2012
434
How it can work with unipolar voltage. Inductor/transformer flux will increase to core saturation with unipolar voltage. Is the coupled inductor an air core inductor?
For the purpose of this question, i removed the coupled inductor from the requirement.
However, Coilcraft informed me that the inductor of interest can accommodate a unipolar oscillation, and will output a unipolar oscillation on the secondary.
 

ronsimpson

Joined Oct 7, 2019
2,987
However, Coilcraft informed me that the inductor of interest can accommodate a unipolar oscillation, and will output a unipolar oscillation on the secondary.
I work with CoilCraft and they know better.

I really don't know that to do. At this point, I have a signal generator that will make the waveform, but at low current. Then connect a Power Amplifier. It will get very hot. Not simple. The power amplifier will have to be DC coupled. No caps on the input or output.
 

BobTPH

Joined Jun 5, 2013
8,809
Why the secrecy? It would really help if we knew what you were trying to do. I, for one, do not help with secret projects.

Plus, you specification do not make any sense. You say the load is 8mΩ. If you put 5V across it, it will pull 625A. If you want 5A through it, the voltage will be 40mV. Sorry, you cannot have 5A at 5V into any load except 1Ω,

Bob
 

Papabravo

Joined Feb 24, 2006
21,159
Here is an idealized design for a unipolar inverter. The AC reference frequency is 3 kHz, but the sampling of the reference sine wave is 72 times that frequency or 216 kHz and the waveform is getting a bit sloppy. At 15 kHz it gets ugly and at 100 kHz it is non existent, with a sample clock at 7.2 MHz. A 100 kHz pure sine wave inverter might be a pipe dream, at least with real components.
1679432043477.png
 
Last edited:

Thread Starter

johnyradio

Joined Oct 26, 2012
434
I work with CoilCraft and they know better.
"with"?

So the Coilcraft technician who told me that their coupled inverter supports a unipolar pulse is wrong?

I thought an inductor responds to change in current, not necessary to alternate between positive and negative.
 

Thread Starter

johnyradio

Joined Oct 26, 2012
434
You say the load is 8mΩ. If you put 5V across it, it will pull 625A. If you want 5A through it, the voltage will be 40mV. Sorry, you cannot have 5A at 5V into any load except 1Ω,
I will have to do a bit of investigating to give info that makes sense. Thx
 

LowQCab

Joined Nov 6, 2012
4,023
At 5-Amps,
a Sine-Wave with a DC offset can easily be done 100% Analog, ( a Heat-Sink will be required ),
but without a definite purpose outlined,
and at least some kind of Load-Specifications, and Performance-Specifications,
it's a total "crap-shoot".

If all of the hanging questions ever get answered,
I'll gladly provide a working Schematic, with part-numbers.
.
.
.
 
Top