Component role in a circuit.

[ktulu]

Hi.
Can anybody explain the exact role of the 100uH inductor in the power section of this welder:

Thanks.

[xavier60]

I don't completely understand the detailed workings. I'd call it an energy recovering snubber.
Lossy snubbers are commonly used with IGBTs to reduce their dissipation caused by tail current during turn off.

I'm referring to the function of the red components

[cur8xgo]

Hi.
Can anybody explain the exact role of the 100uH inductor in the power section of this welder:

Thanks.

Interesting..any chance you could share some more details about the operating voltages/currents/power levels on input and output side?

Soooo..this is a two switch forward converter yes?

Based on the 333 ohm resistance I would imagine that snubber is for the primary inductance, since even at 400V you'd only get about 1.3 amps through it, so not snubbing any welding current transients!

I'm still not sure what that inductor does though. Making a flip-book of current directions during the switching events would probably illuminate it but I'm a bit too lazy for that right now.

I'm going to guess the intent of that 100uh limits peak current to protect all or some of the following: switches, diodes, capacitors. But I won't go as far to say I know if its during switching or during conduction.

[ktulu]

Here is the full schematic:
https://elektrotanya.com/gysmi_161_inverter.pdf/download.html#dl
On Sheet1/6 (last page with schematic) the first box is the power stage that drives the output transformer.
I don't own the welder, but as far as I know, the output current rating is 160A @ 60%, input voltage is 230V.

[xavier60]

Because of the slow IGBT turn off, the idea is to have the current briefly flow through something else, almost always a capacitor.
Because of the whole circuit's symmetry, it's only necessary to take notice of the top half sometimes.

For the top group of capacitors to start passing current through the bottom RH diode at the instant that the top IGBTs start turning off, the capacitors have to be already charged to almost full HVDC, negative on their the top terminals.

This charge is established shortly after the IGBTs are turned on, assuming zero initial charge voltage.
At the instant the IGBTs turn on, full HVDC is applied across the inductor which then does a half cycle of resonance with the capacitors, terminated by the top RH diode.

The 1K resistors and 220pF would be to damp the self resonance of the inductor after it has transferred all of its energy to the capacitors and left open circuited by the diode.

Search for "IGBT tail current"

Added: To help visualizes the circuit operation, imagine that the inductor has a center tap that is tied to half HVDC. And another series diode at the bottom end of the inductor.
Yes, it is a  two switch forward converter.

[bob91343]

If you look at audio amplifiers in general, they all have this inductor.  It's there for stability, to prevent high frequency oscillations.  Without it, watching the output on an oscilloscope you will see little bursts of high frequency oscillations here and there on the wave.  You need the inductor to louse up the gain at the oscillation frequency.

Read a book on this, even an old book.  There are other components often used for similar purposes.

[xavier60]

If you look at audio amplifiers in general, they all have this inductor.  It's there for stability, to prevent high frequency oscillations.  Without it, watching the output on an oscilloscope you will see little bursts of high frequency oscillations here and there on the wave.  You need the inductor to louse up the gain at the oscillation frequency.

Read a book on this, even an old book.  There are other components often used for similar purposes.

The inductor is definitely not there for that reason. The IGBTs are hard switching.

[bob91343]

What is shown isn't the entire circuit.  There is a feedback path back to a low level section of the amplifier.  It's the closed loop that has to be stabilized, not the output stage.

[xavier60]

What is shown isn't the entire circuit.  There is a feedback path back to a low level section of the amplifier.  It's the closed loop that has to be stabilized, not the output stage.

It is not a linear amplifier or anything like one.  It is the primary side switching  of an IGBT inverter welder. The only feedback is to control the drive PWM from the control circuitry on the transformer's output side.
I have explained the real purpose of the inductor and associated circuitry in Reply #4.

Correction: The UC3845D current mode controller IC is on the primary side.

[ktulu]

Yes, it is a  two switch forward converter.

This means, that top and bottom IGBTs are switched in phase.

Because of the slow IGBT turn off, the idea is to have the current briefly flow through something else, almost always a capacitor.

At this stage, the return diodes are not yet conducting?

[xavier60]

"This means, that top and bottom IGBTs are switched in phase."
Yes.
"At this stage, the return diodes are not yet conducting?"
yes. The voltage rise across the IGBTs is very fast to half HVDC because there is more than 1/3 of the output welding current charging the capacitors,  40A or more.
After that, the current will just be the very low magnetizing current.
I figure that the capacitors add about 50ns delay in the rise to half HVDC.