UCC28083: UCC28083D

PSENGINEER Intelux

Intellectual 260 points

Part Number: UCC28083

PROJECT NAME -1 KW DC DC CONVERTER

VIN - 42 TO 58 Vdc ,VOUT - 350Vdc , IOUT- 2.86 A

I had observed non linear mosfet's current waveform. some times it is taking high peak current than other pulses.

I am Attaching Waveform Photo

over 4 years ago

0 John Griffin1 over 4 years ago

TI__Mastermind 34465 points

The variation in value of peak currents in an indication that the gate drive pulse width is changing.

So it is easier to see variations in the peak than in the pulse width.

The reason for this may be due to ripple in the input supply or changes in the load.
Is this causing a problem with the design ?

Regards

John

0 PSENGINEER Intelux over 4 years ago in reply to John Griffin1

Intellectual 260 points

There is Humming Noise from Transformer at 659 Watt Output Power.

Transformer ETD49 WITH P1:S:P2 = 7:63:7 AND Lpri=175uH

MOSFET is heated to 120 degree celsius within 2 min.

Heatsink is used with Dimension length*width*height =(127*19.7*43)mm.

I am attaching Waveform is there any issue with my design

0 John Griffin1 over 4 years ago in reply to PSENGINEER Intelux

TI__Mastermind 34465 points

It looks like both of the push pull FETs are on simultaneously.

One FET has not fully switched off before the alternate FET is turned on.
This will cause excessive currents in the transformer which may lead to core saturation.
The large peak currents will cause audible noise from the magnetic structure.
Please check your gate drive design.
You can slow down the gate turn on and speed up the gate turn off or ensure that there is sufficient dead time between gate drives.

Also check your transformer design.
Ensure that the core flux is much lower than the maximum saturated flux .
Check that the winding is vacuum varnish impregnated or the core is potted.
All of these lead to audible noise issues from the transformer.

Regards

John

0 PSENGINEER Intelux over 4 years ago in reply to John Griffin1

Intellectual 260 points

AFTER INCREASING DEAD TIME BETWEEN TWO MOSFET THE NOISE ISSUE IS RESOLVED BUT MOSFET REACHES >100 DEGREE CELSIUS WITHIN 20 SEC

I ALREADY USED APPROPRIATE HEAT SINK FOR MOSFET

I AM ATTACHING WAVEFORM

0 PSENGINEER Intelux over 4 years ago in reply to PSENGINEER Intelux

Intellectual 260 points

I CHANGED THE CONTROLLER FOR DEAD TIME FROM UCC28083D TO UC3825N

0 John Griffin1 over 4 years ago in reply to PSENGINEER Intelux

TI__Mastermind 34465 points

Hello

I can see from looking at the switching waveforms that the FETs are turning on and off too slow.
There is also an excessive drain voltage overshoot.
Your FETs are getting too hot because of this.
You can reduce the power dissipation or increase the heatsinking if you wish to reduce the component temperature.

Regards

John

0 PSENGINEER Intelux over 4 years ago in reply to John Griffin1

Intellectual 260 points

I INCREASED THE HEAT SINK SIZE AND FORCED COOLING BUT STILL SAME THERMAL PROBLEM

LEAKAGE INDUCTANCE OF TRANSFORMER IS 0.6uH,

SUGGEST ME TECHNIQUE TO REDUCE THIS TURN OFF SPIKE OF MOSFET

0 PSENGINEER Intelux over 4 years ago in reply to PSENGINEER Intelux

Intellectual 260 points

I ALREADY USED SNUBBER

Power management

Power management forum

UCC28083: UCC28083D