UCC28180: TIDA-00443-230V, 900W, PFC with 98% Efficiency for Inverter Fed Drives Reference Design

Lenin Prakash

Part Number: UCC28180
Other Parts Discussed in Thread: TIDA-00443

We have built an PFC converter based on TIDA-00443-230V, 900W, PFC with 98% Efficiency for Inverter Fed Drives Reference Design.

The same circuit has been reproduced without any changes. While testing the circuit in the boost follower mode, we observed the following issues.

Initially the MOSFET failed.

Then the AC input to the PFC is given from an auto transformer . PFC ON and bias is shorted together and connected to an external 15V DC supply.

(Note :For Boost Follower Configuration: Populate R14 and Q4 and set R9 = 21.5k Populate R5 = 0.2 Ohm, this has been followed)

The diode D10 is removed since it energizes the relay at the start and bypasses the 10 ohm resistor. Even after this we found that the when the o/p DC voltage

reaches around 55 Volts DC the gate pulse starts with > 95% duty cycle and the o/p voltage boosted to > 600 V DC.

Following which the MOSFET is removed and the o/p voltage DC ( now same as the i/p DC voltage is increased till 330 V DC) It is found that the gate pulse

remain to be same with a duty cycle of >95 %. Once the voltage is increased beyond. 369 V DC. The gate pulse was 0 for approximately half cycle ( 50 Hz frequency)

and once again 90 % for the remaining duration. Then the Vsense waveform was observed. It is found that the Vsense is not a pure DC and comes with a ripple @ 100 Hz.

It seems whenever the Vsense crosses 5V, the pulse gets shut down and whenever it is less than 5 V, the gate pulse remains at 95 %.

I guess V sense shd be a ripple free voltage and I'm not sure why Vsense is having such a waveform. I have also attached the waveforms of Vsense and Gate pulse

for a voltage of 340 V and 370 V. Kindly help to resolve the issue.

Thanks

Lenin.

over 7 years ago

0 John Griffin1 over 7 years ago

TI__Mastermind 34465 points

Hello Lenin,
This reference design was designed to provide an output voltage of between 250Vdc and 350Vdc so if you are getting over 600Vdc output there is something not quite right in the circuit.
I would disconnect the boost follower section (Q4 in the circuit) and check the output is a stable 390Vdc
It looks like your circuit is in some kind of burst mode.
There will always be some amount of line frequency ripple on the VSENSE input because the loop is slower than the line frequency but this ripple should not be excessive.
Please recheck your circuit and look for any differences between yours and the TIDA-00443
Regards
John

0 Lenin Prakash over 7 years ago in reply to John Griffin1

Intellectual 660 points

Hello John

Thanks for your response. Actually we have not changed anything from the reference design... I have also captured the signal @ I sense which resembles a half rectified waveform... I'm attaching the same below. It is 0V for half cycle of the power frequency, which i guess causing the pulse to remain @ 98 % duty cycle for that half cycle... The load we are using now is a 25 K resistor which is permanently connected...

1. Is there any issues connecting the load @ start ? or shd we need to switch on after some time ?

2. Or connecting such a low current load is an issue ?

Also I'm not able to understand why to allow the max duty cycle till 97 % in this controller for a boost converter..

Pls have a look at the below I sense signal and i there is any issue with this..

As such we haven't changed any from the reference design, however the issue still prevails...

Thanks

Lenin.

0 John Griffin1 over 7 years ago in reply to Lenin Prakash

TI__Mastermind 34465 points

Hello Lenin

The Isense signal should be negative with respect to system ground.

It can never go positive because of the location of the sense resistor.

It should look something like the signal below:

Regards

John

Power management

Power management forum

UCC28180: TIDA-00443-230V, 900W, PFC with 98% Efficiency for Inverter Fed Drives Reference Design