• State TI Thinks Resolved
  • Locked Locked
  • Replies 31 replies
  • Answers 8 answers
  • Subscribers 50 subscribers
  • Views 623 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.

Original question:

INA300: SCH Review: Vpwr = 5V; Itrip = 10mA; I_Max = 100mA

INA300: INA300 schematic review

Gary Teng
Genius 5190 points
Part Number: INA300
Other Parts Discussed in Thread: INA303, INA302, INA301

Hi sir,

 Would you please helping to check my customer's design about INA300? And they want to know the max error% about the result? Thanks for your kindly help.

Best regegards,

Gary Teng

  • 0
    TI__Mastermind 26465 points

    Gary,

    From the limited schematic if looks fine.  The limit resistor does not seem to have a K on the 14 and I assume that is what is required.

    For error to get the best answer that I need some information.  Or you will need to look at our training to determine which error sources matter.   Normally CMRR and PSRR do not play a large part.  Gain error could be dominated by the Rshunt value.  Also you will need to account for the error in you Rlimit resistor.

    https://training.ti.com/ti-precision-labs-current-sense-amplifiers

  • 0
    Prodigy 200 points

    Hi,Sir,

    Could you help to confirm below:

    PoE PD min = 12.95W

    12.95W*0.87=11.26W ----------------PoE module efficiency

    11.26W/[12.5V*(1+5%)]=0.858A-----Current min.

    11.5V*(1-5%)=10.925V-----------------Voltage min.

    10.925V*0.858A=9.373W--------------Power min. 

    Best regegards,

    Zong Wu

  • 0
    Prodigy 200 points

    Hi,Sir,

    At this case,limit resistor Rlimit=10.925V/0.858A=12.73 ohm

    What do you think this Rlimit?

    Best regegards,

    Zong Wu

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir,

    At this case,limit resistor Rlimit=10.925V/0.858A=12.73 ohm

    What do you think this Rlimit?

    Best regegards,

    Zong Wu

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    That is incorrect calculation for Rlimit.  The voltage of your common mode does not matter in the calculation.  I apologize for not seeing that.  In the datasheet there is a table for calculating the Rlimit.  The input scale of the INA300 is from 0 to 250mV differential and your trip point is at 8.58mV.  Assuming your Rsense is 10mΩ and ILimit = 0.858A the calculations are done below.  The Rsense value is a tradeoff of power vs accuracy.  The Rsense value could be increased if noise or increasing accuracy is required.  The 500µV below in the calculations is to account for offset.  As you increase Rsense the Vos is less of a impact.  This is all at a trade off of power and voltage drop accross Rsense.

    ILoadTrip 0.858 A
    Rsense 0.0100 Ω
    Vtrip 0.00858 V
    Rlimit 454 Ω
  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir,

    If I select Rsense is 10mΩ , it will over SPEC.

    RSENSE_MAX Maximum sensing resistor value PRSENSE / IMAX^2  2.5 mΩ.

    Should Rsense select < 2.5 mΩ, right?

    Best regegards,

    Zong Wu

  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir,

    Another question, ALERT is open-drain output.

    Could we set Vs is 4.2V but ALERT is 1.8V pull-up (10k ohm)?

    Will this affect the ALERT output?

    Best regegards,

    Zong Wu

  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir,

    Could you help me to confirm below 

    PRsense
    =Rsense*I^2
    =0.001*0.858*0.858
    =736.2uW (<250mW)

    Rsense
    =PRsense/(I^2)
    =736.2uW/0.858A/0.858A
    =1m ohm(<2.5m ohm)

    Vlimit=0.858A*0.001ohm=858uV

    Rlimit
    =(Vlimit+500uV)/20uA
    =(858uV+500uV)/20uA
    =68 ohm

    Best regegards,

    Zong Wu

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    I would increase your Rsense to the maximum possible to get best signal.  10mΩ would be better than 2.5mΩ if you are worried about accruacy.

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    You are correct that this will be ok and will not be an issue.

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    This will be an issue as you are approaching the offset of the INA300.  The offset is 500µV.  You should increase your Rshunt like mentioned above.  This is a tradeoff on your design of accuracy and error at trigger point or power consumption on the Rsense.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir,

    In the datasheet, how does below value?

    PRSENSE Maximum allowable RSENSE power dissipation 250 mW

    Best regegards,

    Zong Wu

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    This is an example of a Rshunt used with the maximum power rating for that resistor is 250mW.  This is an example.  Use your Rshunt and Rshunt maximum power ratings for these calculations.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir,

    1. In datasheet , Maximum measurement error：5％

        How is this calculated?

    2. At datasheet, 7.3.1.1 Selecting a Current-Sensing Resistor:：Example part

        “Given the total error budget of 5%, up to 3.5% of error is available to be attributed to the internal offset of the device.”

        What is the basis of this "Given the total error budget of 5%" ?

    BR,

    Zong Wu

  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir

    Do you have smaller tolerance value Current-Sense Comparator IC? 

    Like INA260AIPW tolerance 0.1%

    BR,

    Zong Wu

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    1)The 5% is a made up requirement for the example.

    2) This is mentioned to say the error for this example is met.

    • 5% is Maximum Error in this example.
    • 1.5% is Error from Rshunt and ILimit current setting resistor (This is based on the 
    • 2% of Error is from offset (Vos/VsenseMax)*100 = 500µV/25mV * 100 = 2%
    • Total error is 1.5% + 2% = 3.5%

    You can improve the Rshunt value and improve the accuracy but this will not improve the Vos value.  This may not work in all situations as the power may not be acceptable.  For example if my Vsense max was 250mV then the offset would calculate to .2% and you would still need to account for the tolerance of the Rshunt and ILimit resistors.

    For better accuracy with a comparator please look at INA301(Single comparator), INA302 and INA303(Dual Comparator).  This link will give you all the comparator options but  I highlighted the best accuracy options.  For these devices if you use similar calculations you will also need to account for gain error.  In the INA300 this is accounted for in the VOS error.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir 

    Before:

    11.26W/[12.5V*(1+5%)]=0.858A
    11.5V*(1-5%)=10.925V
    10.925V*0.858A=9.373W

    After:

    11.26W/[12.5V*(1+1%)]=0.89A
    11.5V*(1-1%)=11.38V
    11.38V*0.89A=10.13W

    PS.

    Our case Total error:

    Vsense_max

    =Rsense*Imax

    =1.36Ω *0.858A

    =1.17V

    (Vos/Vsense_max)

    =(500uV/1.17V)

    =0.043%    -------Error from offset

    • 1% is Error from Rsense
    • Total error is 1%+0.043%=1.043% (about 1%)

    Is the above calculation correct?

    BR,

    Zong

  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir,

    Error from Rshunt and ILimit current setting resistor

    Rsense error =1%

    RLimit error =1%

    This Initial error=1%+1%=2%

    Is this correct?

    BR,

    Zong

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Zong,

    That is correct with your calculations except your Vsense max is too large.  The largest you can have is 250mV and I would suggest you give some head room and use about 200mV as your trigger point for you Vsense.

    In that case I would go with 200mV = Rsense*0.858A --> Rsense = 0.2/0.858 = .2331Ω.  The above power just on Rsense I think would be too large and the power for this resistor would be 0.1716W but you would get a higher error.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir,

    I use Rsense is 1W 1% at this datasheet,

    So Rsense_max=1W/(0.858A^2)=1.35Ω

    Now I select Rsense=10mΩ < 1.35Ω

    Vsense=Rsense*Imax=8.58mV  < 200mV

    Vsense_max=Rsense_max*Imax=1.35Ω*0.858A=1.15V

    Is this correct?

    BR,

    Zong

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    You should use Rsense~233mΩ explained above.  For you to minimize the offset you should have your trip value as high as possible.  If you are having trouble understanding the tradeoffs please see our training series for current sense amplifiers.

    For your calculations you need to keep your Rsense you select for all calculations.  Your example above still has the Vsense over 250mV which is the max input range for the INA300.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,sir,

    We will select  Rsense=200mV/0.89A=225 mΩ.

    We have other question:

    1.About IEA301, from TI FAE "For these devices if you use similar calcularions you will also need to account for gain error."

    2.Why does IEA301 need to consider Gain Error and IEA300 does not have a note about Gain Error?

    3.Regarding the concept of Total Error of IEA300 and IEA301, for example, if Total Error is 5%, do you think that this is ± 5% for a certain set     

       value? Or do you think it's ± 2.5%? 

    BR,
    Zong Wu
  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir,

    Our schematic has no extra ADC GPIO pin,but INA301 must be out (ADC).

    Could I still use INA301?

    BR,

    Zong

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    1)The reason INA301 has a gain error specification is because there is an analog output that you can measure the output of the analog amplifier.  The INA300 only has a digital output and has no analog output and the error is represented as and offset and includes any error seen by the analog components that would be seen as a gain error.

    2)I believe I covered in #1.

    3)This is +/- 5%

    The INA301 output does not need to connect to an ADC but you can monitor the output if needed.  You could still use the INA301.

  • 0 in reply to Zong Wu
    Prodigy 200 points

    Hi,Sir,

    How about concept of Total Error of INA300 and INA301?

    There are Error_total,Error_initial,Error_available at INA301.

    There is only Error_ available at INA300.

    How do we calculate current sense error (tolerance) at INA300,INA301 respectively ? 

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    The INA300 goes through the example to calculate the error and you have gone through some of the calculations in this post.  Please review that in the datasheet.

    The INA301 has more analog components and the output is analog that is observed you will need to calculate error differently and our training videos goes into depth to help you with this.  You will then need to add the offset of the comparator.  Keep in mind the offset of the comparator is after the gain of the analog component.

    https://training.ti.com/ti-precision-labs-current-sense-amplifiers

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi, Sir ,

    Could I send INA301 alert waveform to you?

    How do I upload photos?

    BR,

    Zong

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    Click insert Image/video/file.  Then click on Upload.  Select picture and click ok. You may have to adjust the dimensions.  Below I inserted some images.

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,Sir

    1. Why Alert_N has many jitters ? 

        It looks like Vsense range to do jitters, how voltage range is this?

    2. Why power on consumption 11.06W Alert_N can't reset system?

        But open AI 9.05W  can reset system.

    Current sense_resin_n_20210510.pptx

    BR,

    Zong

  • 0 in reply to Javier Contreras4
    Prodigy 200 points

    Hi,sir,

    Could you help me review SCH. U707

    yenisei_845_io_bd-v1.1-poe 11.pdf

    Thanks a lot.

    Zong

  • 0 in reply to Zong Wu
    TI__Mastermind 26465 points

    Hello Zong,

    Please let me know where in the schematic you are probing.  The signal names do not match to scope images.  What pull up voltage are you using and what is the pull up voltage.  R779 is a DNI.  Is this correct?

    I would probe on OUT as well to see the signal seen at the comparator.

    For the reset have you verified that Output voltgage on pin 2 is below the comparator output.

    With a high impedance meter you can verify the Limit voltage.