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.

LM317HV: Transient response time of LDO

Part Number: LM317HV
Other Parts Discussed in Thread: LM317, TPS7A85, TPS7A4001, TPS7A94, TPS7A57

Hi LDO team,

I wonder how fast LM317HV can respond to the variation of R2. Customer tries to adjust the voltage of ADJUST pin to vary the output voltage. (like figure 12 in LM317 datasheet)  

1) I believe I can check the BW of LDO as a point of reference when it comes to the transient response time. Are the BW of LM317HV and LM317 same?

2) May I know the bandwidth of LM317(HV)?

3) I notice LM317 has a different structure from other LDO that I'm familiar with - TPS7A85, TPS7A4001, etc.. It look LM317 has an unity gain structure that output node is directly connected to negative node of internal OPAMP. Is this structure has any advantage when it comes to the transient response time?

<LM317 internal structure>

<TPS7A4001 internal structure>

It would be appreciated if you can provide more advice on the difference between two products - LM317 (which is unity gain structure) and TPS7A4001 (which is not) - and on which parameters I should look at in order to check the response time of LDO.

Thank you.

  • Hello,

              I am looking into this, please provide us 2-3 business days to verify if we have the design database for these old parts. Thanks!

    Regards,

    Srikanth

  • Thanks Srikanth for your prompt response. Will wait for your feedback.

  • Hello Ella,

                    At this point it appears to be difficult to locate the design database that can be simulated to generate the bandwidth information. However we can deduce the approximate bandwidth within a range of values with the help of the PSRR graphs. Since the ripple rejection curves appear almost identical, it does look like LM317HV and LM317N have the same bandwidth (in the range of 10KHz-100KHz).

                    Although at first glance it appears that way, please note that the Internal Op Amp (Error Amplifier) in LM317 is not connected in a unity gain configuration. Any perturbation in VOUT (ΔVPER) is fed back to the Error Amplifier not only via OUT pin(where gain of the path is 1) but also via the ADJ pin when the resistor network (RFB TOP & RFB BOT) is connected. In fact for this device the net error signal fed back to the Amplifier would not equal ΔVPER but equal a smaller value: ΔVPER*(RFB TOP/(RFB TOP +RFB BOT)), whereas for devices like TPS7A85 with a conventional architecture, the error signal fed back to the Amplifier would equal ΔVPER*(RFB BOT/(RFB TOP +RFB BOT)). 

                   However I agree with your general understanding, and we do have devices like TPS7A94 and TPS7A57, where the Error Amplifier is always connected in a Unity Gain configuration in the true sense and because of this the bandwidth would indeed be higher. The Line/Load transient response time is a direct measure of the device bandwidth. The time from when the perturbation is introduced to when the LDO output begins to recover towards its nominal value would the easiest measurable metric to compare bandwidth amongst devices. Thanks!

    Regards,

    Srikanth                     

  • Thank you a lot Srikanth,

    May I have one more question? I wonder if this device has internal discharging path.

    The figure below is the customer's concept. FB pin would be ADJ pin of LM317. They're thinking to vary the output voltage by adjusting ADJ pin voltage. If the desired output voltage is stepped up, LDO will source the current to charge the output capacitor. What about the output voltage is stepped down? There should be path to discharge the output capacitor.

    Per I know, most LDOs or DCDCs have internal discharge path with a certain capability so that they can sink current from output node. May I know if there is any internal discharging path in LM317 which is quite old device?

    Thank you for your help!

  • Hello Ella,

                    This device does not have an active output discharge feature. As you guessed, this is a feature found usually only in relatively new devices. Thanks!

    Regards,

    Srikanth