• State
  • Locked Locked
  • Replies 9 replies
  • Subscribers 63 subscribers
  • Views 193 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

LM5175: Does not start with high load for ISNS

Terho Salmijarvi
Prodigy 20 points
Part Number: LM5175
Other Parts Discussed in Thread: LM5176

Hi Team,

 

We have LM5175 which is used for charging a battery, 7 cells 25.9V 3.3A 86Wh.

In our design we want to control both output voltage and output current.

 

For voltage we are biasing feedback (FB) voltage with a voltage from a DAC.

This works fine. We can adjust voltage as we want.

 

We want to limit output current to a value we set by another DAC from 0.1A to 5A.

For current limit we use a 3mR shunt resistor and ISNS inputs.

The voltage over shunt resistor is amplified 100 times with INA280A3.

This voltage is compared to the desired voltage (desired current) from DAC with a differential amplifier.

We have PC software where we can set our desired values.

 

For electronic load this current limit works fine.

Example:

I set desired voltage to 27000mV. I set desired current to 500mA.

Electronic load is set to 200mA. Output is 26.9V. Converter is running.

Electronic load is set to 1A. Output is 6.42V. Converter has stopped.

Electronic load is set back to 200mA. Output is 26.9V. Converter is running again.

 

For battery load the converter does not start correctly if actual current is higher than desired current.

Example with a battery of about 25V. Current will be higher than 500mA if 27V is applied.

I set desired voltage to 27000mV. I set desired current to 500mA.

This picture is after about 16ms and SS has raised above 0.8V.

Current goes up and above 50mV on ISNS+ very fast and SS is pulled down. PWM stops. Current decreases.

SS starts to increase and the procedure continues.

Converter will never start properly.

 

 

If the converter is started and actual current is lower than desired current, then it works/starts.

 

If the converter is working and I thereafter set to 27V and current limit to 500mA then it works.

Here I can hear audible noise of about 3kHz, regulation frequency.

 

I know it don't make sense with my settings like "desired current" instead of actual volt but I hope you get the feeling of how it behaves.

Best Regards

Terho

  • 0
    Prodigy 20 points

    I forgot to tell:

    If I start (EN set to 3.6V) with higher actual current than desired current then converter acts like in pic1.

    If I start (EN set to 3.6V) with disconnected battery and then connect the battery then converter acts like in pic2.

     

    /Terho

  • 0 in reply to Terho Salmijarvi
    TI__Mastermind 42321 points

    Hi Terho,

    thank you for using the E2E forum.

    Can you share the schematic?

    Can you use the quickstart calculator 

    https://www.ti.com/tool/download/LM5175QUICKSTART-CALC

    enter your design data and share the file here again.

    What is your softstart time compared to the reaction time of the average current limiter.

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 20 points

    Hello Stefan,

    Have to ask our customer if we can share schematics. Not done yet.

    I have changed Css from 100nF to 1uF and now it starts with minimum output current. This is good but soft start is now 180ms.

    We don't have any delay in the current measurement and feedback to ISNS. We want to avoid delay here if possible to prevent high current before we can react to it.

    Have put in my component values in Quickstart Tool. I will appreciate if you can check for any errors.

    Best Regards

    Terho

    LM5175 Buck-Boost Quickstart Tool Terho_Working values r9.xlsm

  • 0 in reply to Terho Salmijarvi
    TI__Mastermind 42321 points

    Hi Terho,

    will review the quickstart calculator but need 1-2 days.

    Is there a reason why you have selected the LM5175 and not the LM5176.

    The LM5176 has some improvements, also for the average current limit operation.

    The LM5175 will not turn on Boost High side FET when in current limit during startup as this is controlled by the voltage level of SS pin.

    So the LM5176 might be the better option. The LM5176 is pin compatible to the LM5175 just the Rt resistor needs to be adopted.

    Just to ensure I understood right with the 1uF cap at SS the issue is solved?

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 20 points

    Hi Stefan,

    It is not solved yet. The behaviour regarding “chirps”(pic1) when start is almost solved.

    A solution which worked was to slow down SS. Changed Css from 100n to 1u.

    Sometimes it doesn’t start properly but with a bit larger Css (>1uF) it works.

     

    Design has an adjustable threshold for measuring output current. We set threshold to ISNS+ via diff OP.

    If we want an average current setting of 0,2A, the voltage will be 50mV on ISNS+ for 0.2A out.

    For an average current setting of 2A, the voltage will be 50mV on ISNS+ for 2A out etc.

    INA280A3 outputs 0.3V/A

     

    Now with 1uF Css, regulation is slow.

    I set Vout to 29V. Iout to 200mA. This will give 50mV to ISNS+ when load is 0,2A.

    My battery is about 26V so there will be some current when connecting it to output of charger.

     

    Current rises to 1.5V (5A) at start and decreases to our 0.2A after 6ms. ISNS+ is now 50mV.

    We want to avoid this high “inrush current”. Might affect battery safety circuits?

    I have not looked into LM5176 yet, but do you think it would solve this problem?

    Best Regards

    Terho

  • 0 in reply to Terho Salmijarvi
    TI__Mastermind 42321 points

    Hi Terho,

    not sure if the LM5176 will solve the issue you see here but when directly starting into current limit mode the high side boost MOSFET might never get active and all current will go through the body diode. This might gives you another issue.

    I will check again on the other issue and come back in 1 -2 days.

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    TI__Mastermind 42321 points

    Hi Terho,

    any update for sharing the full schematic (power stage only - not the full application)

    Best regards,

     Stefan

  • 0 in reply to Stefan Schauer
    Prodigy 20 points

    Hi Stefan,

     

    Here is the schematics. I hope this can be solved.

    Best Regards

    Terho

  • 0 in reply to Terho Salmijarvi
    TI__Mastermind 42321 points

    Hi Terho,

    thanks for sharing.

    I am not sure but it might be that the current sense circuit introduces a pole or zero into the compensation loop which leads to instabilities.

    This with increasing the cap on SS and making the current loop much slower gives also an indication to that.

    Analyzing the loop stability in constant current mode would be good.

    You can also try to make the response of the compensation slower with changing the RC to e.g. 1k and 120nF (sorry can not read the designator in the attached images - too low resolution) and check if this improves.

    Best regards,

     Stefan