Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

  • State Resolved
  • Locked Locked
  • Replies 10 replies
  • Answers 1 answer
  • Subscribers 46 subscribers
  • Views 861 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.

INA169: Design help for current monitoring of input to ESC

Juan Ocampo
Intellectual 385 points
Part Number: INA169
Other Parts Discussed in Thread: INA228, INA240, INA290, INA229, INA293

I am currently using the INA169 to measure the current on the high side going into an electronic speed controller (ESC) for an aerial vehicle. Therefore, the current can change very rapidly at any time. The current range is from 0 to 30A. We want the measurements to be fairly accurate throughout the range. With the INA169, the measurement is not very accurate at low currents. Also, we have a microcontroller to read the ADC value, but we could potentially use I2C or some other digital interface. In fact, a digital interface may be better to prevent noise from coupling into the ADC signal.

Here is a bit more information in case it is useful:

Parameter Value Notes
Voltage (V) 3.3  
Rl (ohms)   110,000.00  
Rs (ohms) 0.001  
Rs_tolerance (%) 1%  
Gain (V/V) 110  
Maximum Gain Error (%) 2%  
Iload_min (A) 0.1  
Iload_typical (A) 7  
Iload_max (A) 30  
Vos (V) 0.001 Maximum offset voltage. 1mV is very high
Equation Result Units Notes
Rsh_max = (Vout + Gain)/Iload_max 0.001 Ohms Max current shunt resistor value
Psh= Rsh x Iload_max^2  0.9 Watts Current sense resistor power dissipation @ Full Scale curent
Offset Error at minimum load = Vos / (Rsh x Iload_min) x 100% 1000.00%    
Offset Error at typical load = Vos / (Rsh x Iload_typical) x 100% 14.29%    
Offset Error at maximum load = Vos / (Rsh x Iload_max) x 100% 3.33%    
Total Error at maximum load =sqrt( Offset Error^2 + Rs_tolerance^2+Gain Error^2)  4.01%    

Is this a good choice for this sort of application or are there better options? Personally, it seems like a terrible option because of the offset error at low load(0.1A). Thanks for your help!

  • 0
    Intellectual 385 points

    Also, here is the circuit I am using: 

  • 0 in reply to Juan Ocampo
    TI__Mastermind 30910 points

    Hi Juan,

    It seems the circuit diagram didn’t come through.

    Measuring with accuracy in the full range of 0A-30A is challenging for any device. The trend you noted is true for every current sense amplifier that the % error shoots up at the low end of the current range, though some more so than others. Ultimately the offset of the amplifier starts to limit the system accuracy.

    That said, there are both newer analog and digital current sensors that provide better accuracy. Since the voltage is low at 3.3V, the potential number of candidates is large.

    You may want to explore our online catalog for either an analog current sense amplifier or a digital power monitor.

    The quickest way to get an idea of a device’s performance is to take advantage of the online calculator, located in the product folder. It looks like this (INA169 as an example):

    Regards, Guang

  • 0 in reply to Guang Zhou
    Intellectual 385 points

    Guang, 

    Thanks for the response. The common move voltage for my application is actually around 50.2V max (48V nominal). Also, I was not sure if there was a better current measurement sensor for fast-changing currents such as in this application or if the INA169 was good enough. 

    I took a look at the calculator, but it is basically the same thing that I have. I built an excel spreadsheet that basically plots the error as well.

    Do you have a calculator where you input the parameters and it outputs a suggested part number or something like that? I still don't know what my best option is for this particular application. 

  • 0 in reply to Juan Ocampo
    TI__Mastermind 30910 points

    Hi Juan,

    The best way to select a suitable device is to understand what is required, like you do, and use the filtering function within the online catalog to zoom in on a few candidates. Unfortunately there is not a calculator which spits out a part number based on system requirements. A little bit of manual selection is still required.

    Based on your common mode range requirement, you may look into INA240, INA290 and INA293. These should provide the better accuracy that you’re looking for. For digital power monitors, please look into INA229 or INA228. Check out the minimum conversion time and make sure it satisfies your requirement before going too far, since you mentioned the application is for “fast-changing” currents.

    Regards, Guang

  • 0 in reply to Guang Zhou
    Intellectual 385 points

    Sounds good. I will check those out. 

  • 0 in reply to Guang Zhou
    Intellectual 385 points

    I just took a look at the suggested current monitors. It looks like the digital ones should be fast enough for our application and also they are much more accurate according to the RSS Error charts from your website when using the same parameters:

    Does my analysis above look correct? Is there a reason why there aren't any analog current monitors that are more accurate? All of them have an RSS Error of over 20% error at low currents. Sorry for coming back to this thread that was already closed. Thanks!

  • 0 in reply to Juan Ocampo
    TI__Mastermind 30910 points

    Hi Juan,

    The calculations you’re trying to attach didn’t come through.

    Analog current shunt monitors can be very accurate as well, it is determined by manufacturing cost and the market they serve.

    The few I recommend before, analog or digital, are some of the latest releases with amazingly low offset serving high voltage applications. When it comes to measuring low level current, a device is indeed going to appear more accurate if its offset is 2uV enven though 20uV is already a superior spec.

    Regards, Guang

  • 0 in reply to Guang Zhou
    Intellectual 385 points

    Sorry. Here is the image:

    From these numbers, it looks like I should definitely go with one of the digital modules if I want the best accuracy. 

  • 0 in reply to Juan Ocampo
    TI__Mastermind 30910 points

    Hi Juan,

    These numbers are reasonable. As you can see, the smaller the device offset voltage, the better the accuracy at low current. It is a challenge to measure over a wide current range and maintain accuracy.

    While not rail the output with 30A, increasing the shunt resistor value will improve the % error at 0.1A.

    With high precision systems, the shunt resistor tolerance and layout become important as well, it is something to keep in mind.

    Regards, Guang

  • 0 in reply to Guang Zhou
    Intellectual 385 points

    I appreciate your help. Thank you!