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.

TLV9052: Interfacing load cell to MCU

Part Number: TLV9052
Other Parts Discussed in Thread: PGA900, ADS1115, INA819

Sorry up front if this is in the wrong place in the forum. I need some help choosing a device & design I can use to interface a load cell to a BeagleBone Black Rev C. MCU.    I couldn't post this without choosing a TI Part number and we use the TLV9052 already.

I have a VPG Force Sensors 60001-500L-A1-00F s-type load cell considered for a prototype weighing application on a new piece of stationary agricultural equipment.   The load cell is a Wheatstone bridge type sensor.   We plan to use four of these to measure the weight of the fluid in tank during or process. 
I am an engineer but do not have much experience with load cells interfacing through milivolt outputs.  
The load cell has a full span weight range of 0-500lbf.  It's sensitivity is 3.000mv/V and the maximum excitation is 15VDC and 10VDC is the recommended excitation.  It will be easiest for us to excite the load cell with 12VDC.

We use the BeagleBone Black Rev C microcontroller and my preference, not a hard or fast requirement, is to interface with I2C.    The BeagleBone's I2C voltage range is 0 - 3.3V.

I have attempted to use the Nuvoton NUA7802 ADC on a breadboard to interface this and just can't get that to work.  I think part of the problem is that the load cell needs an excitation voltage higher than the ADC likes.  I haven't blown anything up but the readings are not right. 

I am confident that TI has a good option but haven't been able to identify one.   I know TI will have better documentation, tips and support!  The TLV9052 is not my suggestion. I Just had to choose a part number to be able to post something here.  The PGA900's are going to be hard for us to prototype with on a board unless you have suggestions of mounting methods I could use to handle it manually on a breadboard.

I look forward to hearing suggestions.

WC

  • Hello, WC

    full span weight range of 0-500lbf.  It's sensitivity is 3.000mv/V

    Does this mean output is 3mV/V at 500lbs? Making 36mV maximum output with 12V applied. The common mode is 6000mV, so an instrumentation amplifier is suggested. 

    https://www.ti.com/product-category/amplifiers/instrumentation/overview.html

  •    3mv/V is the sensitivity.   So, at full load, 500lbf, the output is indeed 12V * 3mv/V = 36mv.    And yes the common mode voltage is ~6V.   I'll look at the products you suggested and get back with any questions.

    I sincerely appreciate the rapid reply too!  

  •  After a quick look, the INA2126UA/2K5 looks like it would work.  Since the load cell is excited by 12VDC, then I would power the INA2126UA/2K5 from the same supply.  

    Gain can be set as low as 5V/V with a single resistor.  So, if I excite the load cell with 12VDC at full load I'd get 3.000mv/V *12 = 36mv.  Put in a 1,780 ohm  1% resistor for Rg the gain would be 50.  The amplifier would take this up to 50V*0.036V/V (50*0.036=1.8V) and I could either read it with an ADS1115 (which we already use) over I2C or go directly into an analog input  (max input 1.8V) on the Beagle if four are available.  (We will use 4 load cells per unit.)

    This is for a stationary agricultural system that will likely operate in an unheated structure.  We don't need fast response but it looks like it has it anyway.

    Am I missing anything?

    WC

  • WC,

    Good CMRR is a must unless an unloaded sample data was available to offset this offset error.

    Min CMRR is 74dB. 6V - 74dB is 1.2mV

    1.2mV/36mV * 500lbs = 16.6lb offset error 

  • Good points   I will make sure we can tolerate that error in our system.   16.6lbs is over 7 liters of water so I think we'll need to do better!

  •   I think the INA819IDR looks like it will give us what we need but I'd appreciate your thoughts too.   The load cell will have an excitation voltage of 12VDC and has a 3.000 mv/V sensitivity.    From what I read in the literature, the INA819 has 110db min CMMR so the Input-Referred Common-Mode Error would be just 0.0190mv which on this load cell would be ~0.12 liters/mv.   (load cell output of 13.7 lbf/mv / 2.2046 lbs/liter water density at room temperature = 6.21 liters/mv *0.019 = ~0.12 liter of ~120ml which is good enough for our purposes about 0.66% error at our lowest batch size)   Ambient temperature swing possibly 10 C would introduce a 4uV shift which we can consider negligible.  Maximum input offset is 35 uV, which again, against 0-36mv would be negligible.

    Just a few design questions remaining.

    1) can i power the IN819IDR from the same 12V supply as the excitation voltage?  Essentially -Excitation to the sensor connected to -Vs and + excitation connected to +Vs and have -Vs also connected to the ground of the BeagleBone Black which will eventually read the output?

    2) Is there really any need for a second op amp other than the desire to read this voltage with I2C?

    Any final thoughts/comments?

  • WC,

    I suggest using orange "Ask a related question" with INA819 as the part number to get final check.

    1) yes

    2) IC2 will likely need another power supply voltage. 

  •  I posted the related question and a schematic.   The I2C will get its power from the BeagleBone Black @ 3.3 V to match the Beagle's I2C bus voltage.