Clamping of Mosfet Drain to Source Voltage.

Thread Starter

dharanimoka

Joined Jan 29, 2018
18
I'm simulating a Current Fed dual inductance half bridge converter. The voltage across the mosfet on the primary side is not getting clamped to the required voltage level. I was thinking that using a passive clamper circuit might help. May I have anymore suggestions.

I have uploaded an attachment of the gate pulse and the voltage across the switch when the switch is turned off.
 

Attachments

ericgibbs

Joined Jan 29, 2010
18,766
hi dh,
Welcome to AAC.
Do you have a circuit diagram of the converter that you could post, it would make it easier to advise you.
E
 

ericgibbs

Joined Jan 29, 2010
18,766
hi,
It looks as though you are using LTSpice.?
If so, please post your LTS asc file, so that we can run the simulation.
E
 

ericgibbs

Joined Jan 29, 2010
18,766
hi dh,
I have run the sim , using UF4007 diodes across L1 and L2, runs OK and clamps.
Try the UF4007's.
E

EDIT:
It ran OK for 2 sims then failed.!!! I will keep looking.
 
Last edited:

ericgibbs

Joined Jan 29, 2010
18,766
hi,
What is the Vout design voltage.
E

EDIT:
Where did you get the circuit design.?
You have 6 separate Gate driving sources.

E
 

ericgibbs

Joined Jan 29, 2010
18,766
hi,
I got to 189Vout, using 100R in series with a 10nF cap across L1 and L2, removed the UF4007.
It takes ages to run a simulation.
E
 

ebp

Joined Feb 8, 2018
2,332
Don't mess around trying to clamp or dissipate energy until you understand the intent of the circuit and where that energy should be going.

First, you simply can't go putting freewheeling diodes across inductors that are directly in the power path and very clearly there for some specific purpose. That much is obvious without looking at the paper.

I had a very brief glance at the paper. The circuit is describe as zero voltage switched / zero current switched. The implication of this is that the timing of the switching is critical so that it happens at zero voltage/zero current, which typically implies that energy is being moved from one storage element to another. The high voltage spikes, if there is actually any energy being switched, suggest that the switching timing is incorrect. It is never sufficient to look only at voltages in these circuits. You must also look at currents.

While it is necessary to understanding the circuit, the IEEE would likely stomp all over you for posting a copyright paper.
 

Bordodynov

Joined May 20, 2015
3,177
See
I replaced the first two transistors with more suitable ones.Those transistors that you used had an acceptable voltage of 450 volts, but in the model this is not reflected and when simulating the voltage at the drain is more than 1000 volts.I'll fix it (I'll supplement the model).You have unsuccessfully chosen the transmission factor of the transformer.This forces the transistors (input) to work with a high voltage.To limit the large peaks (bursts) of the voltage, I added two capacitors.I advise you to change the transformer.
 

Attachments

ebp

Joined Feb 8, 2018
2,332
Once again, this is a circuit that was devised for a very specific purpose and to operate in a particular way. To mess with the circuit is to change the intent. As I understand the paper from a cursory view, the intent is that proper management of switching precludes the need for added junk to suppress high voltages that wouldn't be there if switching were properly managed. This is a current fed topology and a novel one at that.

If the left hand side is input, the right hand side is a rectifier and the timing must correspond with what is going on on the left. You cannot go at this piecemeal. The whole circuit has to work as a unit.

The title of the paper kind of supplies a whole lot of important clues (emphasis mine):
Analysis, Design, and Experimental Results of Novel Snubberless Bidirectional Naturally Clamped ZCS/ZVS Current-Fed Half-Bridge DC/DC Converter for Fuel Cell Vehicles
 
Top