UCC28911: UC29811 operation with secondary side voltage detection by using an optocoupler

Jonathan Geronga

Part Number: UCC28911
Other Parts Discussed in Thread: UCC28740

Hi Team,

I would like to get your assistance regarding the design of our customer using UC29811.

- The customer would like to confirm if UC29811 can operate with secondary side voltage detection by using an optocoupler.

In the part of their schematic below, they find that it has ~1V voltage swing that looks like ripple at the output. Also, the circuit seems to reset when EMI is present on the AC input voltage. Do you have any thoughts on how to eliminate these issues?

Only that the schematic differs from the reference design in that part of the input filtering is removed to accommodate measuring the voltage between R3 and R8. They are looking for a solution that would not require that inductor and resistor that were removed.

Looking forward to your input.

Regards,

Jonathan

over 3 years ago

0 Pye Naing over 3 years ago

TI__Prodigy 590 points

Hello,

I have assigned this question to an expert on the part. He should give you a response soon.

0 Benjamin Lough over 3 years ago in reply to Pye Naing

TI__Mastermind 26225 points

Hi Jonathan,

The VS pin is designed to look at the divided down AUX winding and the UCC28911 cannot be combined with an opto-coupler for secondary side regulation unfortunately. We have other flyback controllers such as UCC28740 that might be suitable if your customer prefers secondary side regulation.

When EMI is present on the input, could you share what is the exact test condition your customer is performing? Is this a intentional injection of noise on the AC input?

Best Regards,

Ben Lough

0 Jonathan Geronga over 3 years ago in reply to Benjamin Lough

TI__Expert 7930 points

Hello Ben,

Thank you for your support in this thread.

Regarding your question with the exact test condition the customer is performing:

He has injected noise both intentionally and unintentionally.
Intentionally: Inject noise per IEC 61000-4-5 (1000V) OR just connecting a small AC desk fan to the power strip that the DUT is plugged into and clicking the fan on and off.
Unintentionally: When there is some welding being done in our building it will reset the UC28911. Some 50-hz to 60-hz power supplies will trigger it to reset as well.

Please let me know your comment on the customer provided information above.

Regards,

Jonathan

0 Benjamin Lough over 3 years ago in reply to Jonathan Geronga

TI__Mastermind 26225 points

Hi Jonathan,

When you say the device resets, do you mean the device is going into fault where the VDD cap is allowed to discharge to VDDoff and then charged back up to VDDon?

My suggestion would be to take a look at the anode of D5 to get an idea of what the VS pin is seeing when the fan is turned on and off. Perhaps common mode noise is creating some issues for the AUX winding sensing. If this is the case, adding an EMI choke to the AC side of the diode bridge may help.

Best Regards,

Ben Lough

0 Jonathan Geronga over 3 years ago in reply to Benjamin Lough

TI__Expert 7930 points

Hi Ben,

Here's the response of the customer after I shared your answer:

It does not appear that the Vdd cap ever discharged.
I've attached some oscilloscope outputs (The blue line is attached to D5's anode or to Vdd. yellow line is the output ~5V, green line is a current clamp on the output):

D5 Anode during normal operation:

Vdd during normal operation:

Vdd during EMI event:

Vdd during another EMI event:

D5 anode during an EMI event:

D5 anode during another EMI event (severe enough to bring down the 5V line for a few hundred us):

It does look like noise is getting to the VS pin. They have tried adding an EMI choke to the AC side of the diode bridge with little luck (they have actually tried a number of common-mode chokes, cable ferrites, x/y capacitors, and larger line input filters.)

Do you have any other suggestions to mitigate the noise?

Please let me know your inputs.

Regards,

Jonathan