• State Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 2 answers
  • Subscribers 67 subscribers
  • Views 560 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS23752: the threshold between PWM and VFO on output loading

Isaac Wang
Prodigy 20 points
Part Number: TPS23752
Other Parts Discussed in Thread: TPS23751, TLV431A,

From TPS23752 datasheet "8.2.2.7 Setting the PWM-VFO Threshold using the SRT pin" as below.

How could I know the threshold between PWM and VFO on output loading?

Could I get the formula of threshold between PWM and VFO on output loading?

8.2.2.7 Setting the PWM-VFO Threshold using the SRT pin
The TPS23751 and TPS23752 internally compares modified voltages at the SRT and CTL pins to determine the
operating mode. The designer should consider the light load operating point (considering the value of VCTL)
where synchronous rectifier (M2 in Figure 32) gate drive and switching losses nearly match conduction losses of
the rectifier diode (DOUT in Figure 32). Typically, the designer characterizes circuit efficiency, output load, and
control pin (VCTL) voltage and then select the transition point. Both VFO → PWM (occurs at higher load current
due to natural hysteresis) and PWM → VFO (occurs at slightly lower load current) transitions should be
considered when choosing the VSRT setpoint. As an example:
1. Assume that the desired efficiency transition threshold occurs at 18% of full load and VCTL = 2.0 V

2. Determine where to set VSRT.
(2)
3. Set VSRT using a voltage divider from VB to ARTN as shown in Figure 32.
4. Choose RSRT1 = 100 kΩ and then calculate RSRT2 as follows:
(3)
5. Select 11 kΩ for RSRT2.

  • 0
    TI__Mastermind 37121 points

    Hi Isaac,

    Thanks for reaching out!

    To enter the PWM mode from VFO mode the CTL needs to be higher than VCTL_VFO (ignoring the hysteresis). This value may changes according to different V_SRT values.

    For its relationship with load, ignoring AC signal, V_CTL = VB - R45 * [(Vout - Vca)/R47 * CTR]. In steady state, Vca normally increase with higher load power.

    Best regards,

    Diang

  • 0 in reply to Diang Xing
    Prodigy 20 points

    Could I know Vca voltage?

  • 0 in reply to Isaac Wang
    TI__Mastermind 37121 points

    Hi Isaac,

    Sorry for the late response. 

    Vca normally increase with higher load power. But I am not quite sure about the Vca voltage. I have forward your question to the PDS team and hope they are more familiar with TLV431A.

    Here is the test data which may help:

    - when at 1W load, V_OUT = 5.05V, V_AU7 = 4.37V, V_CTL = 1.88V in measurement. If we assume CTR = 80%, by calculation V_CTL(cal) = 1.81V

    - when at 25W load, V_OUT = 5.03V, V_AU7 = 4.50V, V_CTL = 2.76V in measurement. If we assume CTR = 80%, by calculation V_CTL(cal) = 2.54V

     

    Best regards,

    Diang

  • 0 in reply to Diang Xing
    Prodigy 20 points

    Thanks for your reply. Could I know how to calculate Vca in different load so that the threshold between PWM and VFO can be clarified?

    In TPS23752EVM-145: Evaluation Module for TPS23752 spec Figure 2. TPS23752EVM-145 Schematic (2), V_SRT=0.723V get VCTL_VFO=2.15V.

    What's output load when VCTL_VFO=2.15V?

  • 0 in reply to Isaac Wang
    TI__Mastermind 37121 points

    Hi Isaac,

    Thanks for your reply. We need to know the TLV431A's cathode voltage for this calculation as you mentioned. I will reopen your question and forward your question to the PMIC team and hope they are more familiar with TLV431A.

    Please do not check "solved" until all of your questions have been solved. This will close your question automatically.

    Best regards,

    Diang

  • 0 in reply to Diang Xing
    TI__Genius 11401 points

    Hi Diang 

    Alex is running couple of experiments . He will get back to you. 

    Regards

    Trailokya 

  • 0 in reply to Diang Xing
    TI__Prodigy 170 points

    Hello Diang,

    For this flyback control circuit, the cathode voltage will vary with the Vout voltage. Typically how this circuit work, is when Vout rises above its desired threshold, the current through the optocoupler will increase, and the flyback controller will utilize this signal to change the PWM and thus correct Vout.  I ran a PSpice simulation for your circuit (below) that can give you a rough estimate of what cathode voltage to expect. In the past I've used PSpice for a similar circuit and found the measured value in the lab to be close to that on PSpice (however not quite exact).

    This simulation shows that when the Vout (voltage above the optocoupler) value oscillates above its set value by 0.1V, the cathode voltage drops by as much as 1V. This means the voltage difference across the diode (and 604Ohm resistor) increases by 1.1V, and the current does along with it. An increase in the current through the optocoupler will increase the current on the primary side (CTL). Also, above it was assumed that the CTR ratio of the optocoupler was 80%, and while this is possible, it should be noted that this value varies greatly with temperature and biasing current. From what I've seen it would be much lower (maybe around 10%-20%), however you should look through the datasheet to confirm this.

    Also, a little of background on this circuit topology that may be helpful: R48 is utilized to bias the voltage reference so that it's stable throughout the operation, and C37 exists to add a zero to the circuit to extend the bandwidth (better stability across a wider frequency range).

    In conclusion, there isn't a formula I know of that can provide you with Vka as a function of output load, however the relationship between the output voltage and Vka shown above may be useful. 

    I'm sorry I couldn't be of my assistance with your question. I'm going to consult internally to see if anyone else can help with this.

    Thanks,

    Alex

  • 0 in reply to Alex Noggle
    TI__Expert 8035 points

    Hi Alex,

    Do you have further update for this? now customer is face some issue on PWM and VFO switching.

    Regards,

    Mark

  • 0
    TI__Prodigy 170 points

    Hello Mark, 

    Let's try to figure this out on email, I can loop in some people who are more experienced than myself there.

    Thanks,

    Alex