UCC256402: external shutdown

Hi Kurumi,

Kurumi Hasegawa said:

[Q1] To enable external shutdown, I believe FBreplica should be less than BMT (FBLessThanBMT). How can we determine this BMT value? Is there a way to measure this value, or can we calculate it numerically?

[Q2] To enable external shutdown, FBreplica should be lower than SSreplica as well. How this SSreplica determined? Is it 3.5V fixed continuously, or the min is fix to be 3.5V?

[Q3] Related to Q2, does BLK-voltage affect SS-voltage? From functional block diagram, it looks independent.

 

1. You need to pull down the BLK pin for external shut down. Pulling down FB pin is not recommended. External resistances connected at the LL/SS pin and BW pin determine the BMTL/BMTH thresholds. Please see the datasheet for additional details

2. After soft start, ss pin voltage will be at 3.5V which is higher than BMTL/BMTH thresholds. So No need to pull down SS pin voltage for external shut down

3. BLK pin voltage doesnt impact the SS pin voltage. As soon as BLK pin voltage reached BLK start threshold, soft start will be initiated.

Regards

Manikanta P

TI_To the person in charge

Nice to meet you, my name is Kurata from Hinox.

Thank you very much for answering the question of Japanese TI "Kurumi Hasegawa".

Next, I would like to ask you the following additional questions directly.

Thank you for your support.

< I recognized that the summary of the previous answer is as follows. ＞

Item 1 does not recommend pulling down the FB pin.

Item 2 is that the SS pin voltage is fixed at 3.5V and there is no need to pull this pin down with an external shutdown.

Item 3 is BLK voltage does not affect SS pin voltage.

<Additional question>

I understand the above, but I have another question.

<Question 1>

Why is it not recommended to pull down the FB pin?

The EVM application of this device has a circuit example of remote ON/OFF by this FB terminal operation.

Learning from this, I am thinking of stopping the device output by sinking the FB pin.

The reason is that stopping the operation by pulling down the BLK input does not satisfy our power supply system function using this device.

<Question 2>

If the FB terminal can be used for remote stop operation, is it necessary to clear the BMTL/BMTH threshold as a stop condition?

If so, is the threshold derived from the value of the external resistors connected to the LL/SS and BW pins?

You commented that the calculation of the threshold is based on the data sheet, but from which formula can it be calculated specifically?

<Question 3>

With our prototype EVM board, the device was stopped by a sink operation with an FB voltage of about 3V.

However, based on the theoretical derivation based on the calculation results, if this stop threshold is even lower, is it possible to stop by series sink operation of thyristor + diode (sink voltage operating value ≈ 1.3 V)?

Akira Kurata said:

Why is it not recommended to pull down the FB pin?

The EVM application of this device has a circuit example of remote ON/OFF by this FB terminal operation.

Learning from this, I am thinking of stopping the device output by sinking the FB pin.

The reason is that stopping the operation by pulling down the BLK input does not satisfy our power supply system function using this device.

After the FB pin released, the controller might trigger the OCP since reference voltage is different from the current output voltage of the converter. Where as if you release the BLK pin, the controller will go into soft start mode where controller currents will be gradually increasing.

Akira Kurata said:

If the FB terminal can be used for remote stop operation, is it necessary to clear the BMTL/BMTH threshold as a stop condition?

If so, is the threshold derived from the value of the external resistors connected to the LL/SS and BW pins?

You commented that the calculation of the threshold is based on the data sheet, but from which formula can it be calculated specifically?

Fbreplica signal which is inside the controller has to go below BMTL threshold for the switching pulses to be off.

Please see the section 2.6.3 of the app note for calculating BMTH and BMTL thresholds: https://www.ti.com/lit/pdf/sluaaj7

Akira Kurata said:

With our prototype EVM board, the device was stopped by a sink operation with an FB voltage of about 3V.

However, based on the theoretical derivation based on the calculation results, if this stop threshold is even lower, is it possible to stop by series sink operation of thyristor + diode (sink voltage operating value ≈ 1.3 V)?

Its not voltage at the FB pin that decides turn off. Its the current thats being pulled from the FB pin that will decide the fbreplica magnitude. Please go through section 2.3 of the app note for clear understanding: https://www.ti.com/lit/pdf/sluaaj7

Regards

Manikanta P

Manikanta P

This is Kurata from Hinox.
thank you for your reply.
I understand the content.

In our usage, the signal that sinks the FB terminal is only used to stop the output.
Restarting will be a pull-up start from the BLK pin input by turning on the IC power supply again, so there is no need to worry about an OCP trigger due to his FB pin opening after stopping.
Also, I know the FB terminal is controlled by a current sink, but I wanted to know at what volt the voltage would stop when the FB terminal finally pulled.
According to the answer, in section 2.6.3 of the application note he states that the minimum BMT threshold is 0.2V if the upper voltage divider resistor R2 connected to the LL/SS pin is removed. From this description, I understand that logically, if the FB pin is pulled down to a voltage lower than this, all the output stop conditions of the device will be cleared and OK.
Therefore, the operation during this abnormal operation is changed to a circuit that can pull down the FB pin to 0.2V or less.
<Re-question 1>
Also, some answers say "ZCS protection is disabled if Fbreplica is lower than his BMTH", but I don't think there is any feedback control that would make this voltage he below 0.2V in normal operation. Therefore, during normal operation he believes there are no conditions to disable ZCS. This time, setting his FB pin pull-down by IC external operation to the stop level that interrupts his BMT threshold disables his ZCS operation inside the device. However, since the device output shifts to stop mode at this time, it is judged that disabling the ZCS operation does not affect operation.
Is there a problem with this idea?
<Re-question 2>
If an error occurs in normal operation, what is the response time from signal input operation to device output stop in the two methods of "stop operation from the BLK pin" and "stop operation by FB pin sink"?
Also, which one reacts faster?
Also, how many μsec is the difference in response time between these two actions?
Please answer with specific numbers.
Please tell me again.

Hinox Kurata

Akira Kurata said:

Also, I know the FB terminal is controlled by a current sink, but I wanted to know at what volt the voltage would stop when the FB terminal finally pulled.

When you Pull the current more than( IFB source plus clamp current ), vfbreplica will be below 0.2. This would make the controller stop switching.

Akira Kurata said:

Also, some answers say "ZCS protection is disabled if Fbreplica is lower than his BMTH", but I don't think there is any feedback control that would make this voltage he below 0.2V in normal operation. Therefore, during normal operation he believes there are no conditions to disable ZCS. This time, setting his FB pin pull-down by IC external operation to the stop level that interrupts his BMT threshold disables his ZCS operation inside the device. However, since the device output shifts to stop mode at this time, it is judged that disabling the ZCS operation does not affect operation.
Is there a problem with this idea?

When you stop switching, you dont need to worry about ZCS.

Akira Kurata said:

f an error occurs in normal operation, what is the response time from signal input operation to device output stop in the two methods of "stop operation from the BLK pin" and "stop operation by FB pin sink"?
Also, which one reacts faster?
Also, how many μsec is the difference in response time between these two actions?

Both reacts faster. I would assume there are internal delays of the comparators which will be very very small. We don't have this data in Datasheet. So, I cant provide this data. 

Regards

Maniaknta P

<Inquiries about UCC256402 operation problems>　　　　　　　　　　　　　2023.4.07 (Friday)

Manikanta P

This is Kurata from Hinox.

Thank you for your reply the other day.

Another problem occurred in UCC256402, so I will contact you below.

<Symptoms>

In the application circuit using UCC256402, the IC malfunctioned during the following process when the input power supply was cut off. 　

① When the input power supply is stopped, the output pulse stops when the BLK input voltage is below the threshold due to the drop in the bulk voltage.

② After that, Vcc decreases. ⇒ Rvcc output stopped. ⇒Vcc voltage recovery ⇒Restart when reaching 26V. ⇒ Rvcc restart.

　⇒ Rvcc's external connection circuit malfunctions due to hunting.

③ After that, repeat hunting three times.

④ The Vcc voltage drops further, the HV circuit does not start, and the Vcc capacitor discharges. All circuits stop.

<Questions about the above measures>

① Is there a simple way to prevent Rvcc from hunting?

② Is this problem inevitable in the internal circuit configuration of UCC256402, and is it necessary to change the device as a countermeasure?

③ Currently, is there any way to prevent Rvcc voltage hunting if it is difficult to deal with it with the relevant IC?UCC256402_Waveform of operation failure inquiry.pdf

Hi,

Akira Kurata said:

s there a simple way to prevent Rvcc from hunting?

If the controller is goes into fault state, RVCC will be zero. There is no way you can change that behavior.

Also, VCC voltage needs to be at least 13V.  So, you need to make sure VCC is within the recommend range.

Could you let me know how the HV pin is connected to?

Regards

Manikanta P

Manikanta P

This is Kurata from Hinox.
Thank you for answers.
<Re-explanation of the current problem>
The LLC input supplies the bulk capacitor voltage (PFC output) boosted by the PFC to the HV terminal through a resistor of approximately 5kΩ.
The problem this time is the behavior of the Rvcc pin voltage when the input power supply is cut off.
When the input voltage stops, the voltage of the bulk capacitor gradually discharges and drops to 0V, but in the middle of the operation, Vcc drops from the steady DC 15V and falls below the lower operating limit (threshold 13V).
At this time, I recognized that the HV circuit restarted and increased from 13V to 26V, and the temporarily stopped Rvcc was output again, causing hunting.
<Question again>
Is it possible to create an application that avoids the above phenomenon (does not restart the Rvcc output)?
Also, as a possible countermeasure, is there no other way to deal with it than to replace it with an external power supply driven UCC256403 that does not have a built-in HV circuit?
Furthermore, when replacing with UCC256403, is it possible to select continuous mode without using burst mode in static operation at light load as with UCC256402?

Could you share your schematic? I will answer your questions after I receive the schematic file from you.

Regards

Manikanta P

Manikanta P

This is Kurata from Hinox.

thank you for your reply.

Attached is the requested UCC256402 circuit diagram (for submission).

Thank you for your consideration.

New power supply(TI提出用).2023.0412.pdf