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.

WEBENCH® Tools/LM5121-Q1: Minimum Voltage on Webench

Part Number: LM5121-Q1
Other Parts Discussed in Thread: LM5121, LM5155, BQ24640

Tool/software: WEBENCH® Design Tools

Hi,

I would like to use LM5121/-Q1 in my design and use Webench to confirm. My design will have external Vin supply of 5V and it should enable me to bring down minimum  input to 3V. But the webench tool only allows 4.5V minimum. Can you advise solution?

Thanks,

Emman

  • Hi Emmanuel,

    According to datasheet, LM5121/-Q1 need a minimum of 4.5 V for the device to get through start up phase and attain steady state. After attaining steady state Vin can go as low as 3 V until the point where it goes into shut down mode. Once the device goes to shut down mode, you'll need a supply of 4.5 V to restart the device again.  If your input source drops to three volts and if the device goes into shut down mode (due to transients or any abnormality), It may not be able to restart after that. Hence, Webench limits users from creating designs below 4.5 V Vin.

    I hope this answers your question. Please let us know  if you have further queries.

    Regards,

    Vishwanath

  • Hi Vishwanath,

    Thank you for attending to my question. Like I said, I intend to power Vin via external source (low power boost) and does not expect it to go below 5V. If the datasheet specifically mentioned 3V minimum input (for boost, not Vin), I expect to be possible to be used down to 3V. I think it would help to reduce the 4.5V limit so the tool may still be used keeping in mind the Vin requirements.

    But of course I only say this because I need it :). Just checking if you would consider.

    But in case it is not possible, is my plan on supplying Vin externally going to work so I can boost the high current 3V? Am I right in assuming that voltage monitoring is done only on the Vin part and not on the other pins? If so, I will just manually compute the values for 3V input. (for boost, not Vin)

    Thanks,

    Emman

  • Hi Emmanuel,

    Changing Webench limits may create bad designs on webench for the end users.

    I'll move this thread to the applications design team who can help you with your specific concern.

    Regards,

    Vishwanath

  • Hi Emmanuel,

    Thanks for considering the usage of the LM5121 in your design. I will answer your question in two parts.

    1) Is my plan on supplying Vin externally going to work so I can boost the high current 3V? 

    Yes, you can supply VIN externally so you can boost the 3V input voltage. 

    However, you will need to account for the CSP/CSN common-mode voltage, which has a minimum of 3.0V, so if your input voltage deviates below 3.0V, successful operation cannot be guaranteed.

    2) Am I right in assuming that voltage monitoring is done only on the Vin part and not on the other pins?

    VIN pin on the chip is the input to the VCC regulator. The reason the datasheet specifies a voltage of at least 4.5V at the VIN of the chip is that the VCC UVLO is 4.1V. 

    Furthermore, depending on application, voltage monitoring is also done on the UVLO pin. You will need to size those resistors accurately so that the part will turn on/off at the correct voltage. 

    Hope this helps. 

    Thanks,

    Richard 

  • Hi Richard,

    Thank you for attending to my concern. Regarding your answer on number 1, it will put my design on the borderline of operation. Do you have alternative part that has lower common mode voltage on that portion?

    Can you advise better controller for my design? My basic specifications is below

    Vout = 9V 6A

    Vin = 3V - 4.2V

    Thanks again,

    Emman

  • Hi Emmanuel,

    If non-synchronous controllers can be used in your design, the LM5155 can be a suitable controller since it does not have a common-mode voltage for the current sense amplifier. Bias can be powered externally with your boost input voltage.

    Thanks,

    Richard

  • Hi Richard,

    Thank you for your LM5155 suggestion. I made a prototype following the layout in datasheet/evm with values taken from the LM5155 calculator tool. But on my actual test the output voltage drops at just ~2.5A output. I have attached the schematic, pcb layout and LM5155 spreadsheet calculator. Can you check if I may have missed something?

    Thank you,

    Emman

  • Hi Emman,

    Thanks for using the LM5155 in your design! I've noted some comments below.

    SCHEMATIC

    1) According to the LM5155 datasheet, if you connect VCC to BIAS, you can power bias in the operating region from 2.97V - 16V. Otherwise, you will need to supply a voltage greater than or equal to 3.5V. In your case, did the output drop when your bias pin voltage was lower than 3.5V?

    2) The soft-start capacitance is quite big, but it should be ok.

    3) Add a pull-up resistor from Pgood to a system rail. 

    4) UVLO voltage divider is ok for nominal value, but I would try to size different values so that you can cover the range of the threshold voltage. See step 6 of the LM5155 calculation tool. The calculated value is 40.2 KOhm, but I'm not sure why you chose 20 KOhm.

    LAYOUT

    1) layout looks ok in regards to the switch node closeness. The gate driver loop could be closer, but seems ok. I will need scope shots to better assess.

    Scopeshots


    1) Does the output drop only at 2.5A? Or does it drop at loads higher than that? In either case, can you take a scope shot of the SW, COMP, UVLO, and Vout?

    Thanks,

    Richard

  • Hi Richard,

    My Vcc is actually connected to 5v external source. So i did not find the need to connect bias to vcc.

    I was not able to update but the uvlo resistors are actually both 40.2k.

    After fixing the comp resistors and having freshly charged batteries, am able to output 8.75V/4.5A at 3.3V input. I will check in detail if my problem occurs at near 3V and will give you update.

    Thank you,

    Emman

  • Hi Emmanuel,

    Do you have any updates?

    Thanks,

    Richard

  • Hi Richard,

    My final configuration is 8V 3A output at 10V input. The input is now 10V 4.5A. After compensation adjustsment on 3.3V-10V LM5155 boost, the circuit is now stable and I am getting around 25W at the output. I can also get 2.75A short circuit current. I also temporarily bypassed the high RDSon of the P-Ch anti reverse polarity FET.

    I am now using it to charge my 133F capacitor. I think it is normal that when the capacitor is reaching the charger's output voltage, the charging current gets lower and lower. Can you advise if there is a way to have the maximum current even at close to top up voltage?

    Thank you very much for your assistance.

    Regards,

    Emman

  • Hi Emman,

    Thanks for the update. 

    Is there a specific reason why you would want to keep charging the capacitor at maximum current close to the desired output voltage? I would be wary of overcharging the capacitor and then damaging it. There is a way, however, to have constant-current charging alongside constant-voltage monitoring.

    One method for having constant current charging is to add a constant-current (CC) loop alongside the constant-voltage (CV) loop, which is provided with the FB pin. So what I would recommend is to convert the charging current to a voltage using a device such as a current-sense amplifier, and then feed that to a op-amp integrator with the anode of the diode connected between the output of the op-amp and COMP. 

    In this way, if the voltage of the capacitor is reaching towards the final value, the feedback resistor divider will control the loop, since the current signal is small enough such that the diode is reversed biased, thus removing any effect of the charging current from the control loop. During charging, the output of the integrator will end up forward-biasing the diode so that COMP is set to a value that will lead to a constant-current charging the capacitor. 

    While this method does not exactly get you to charge maximum current even at close to the output voltage, you are able to use the charging current to control the loop.

    Thanks,

    Richard

  • Hi Richard,

    My application requires that the capacitor is at top up voltage immediately after discharge.

    Do you mean aside from the built in current sense of BQ24640, I will tap on the same current sense resistor to implement constant current loop alongside CV loop? where is this COMP that you mentioned?

    Thank you,

    Emman

  • Hi Emmanuel,

    I meant in that from the 5155 perspective, if you were to implement a CC loop, you should populate a current sense resistor in series to the BQ charger, and create the current sense amplifier, to integrator with diode circuit and connect that to the comp of the LM5155. 

    However, since you are using the BQ24640, I do not think you will need to you this implementation I suggested. I recommend asking the BQ24640 engineers for information on how to have maximum current when you are reaching the expected output voltage of the capacitor since the LM5155 will boost to the correct voltage that you will need. 

    Thanks,

    Richard