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.

INA826EVM: Compatibility with 4-wire Load Cell and Connection Confirmation

Part Number: INA826EVM
Other Parts Discussed in Thread: INA826, , TINA-TI

Dear All, 

My query was regarding the use of INA826EVM (INA826 Evaluation Board) with a 4-wire load cell, and I just wanted to check whether the following connections are correct, as I wanted to simulate the same before purchasing the module.

System Description:

The complete circuit diagram is shown here.

(Figure 1)

The load cell used contains 4 strain gauges at 350 ohms each, the positive and negative excitation to the load cell are between 5V to 7.5V at a sensitivity of 2.0mV/V and 5 N-m is the maximum torque. 

For my application, the output voltage of the load cell is approximately 2 mV (at +5V, -5V excitation), and to connect the same to an ADC, I am attempting to use an INA826 evaluation module set at R_gain = 100 ohms to obtain ~0.99V at VO.

I have also verified the same with TINA-TI software, and the results are shown below. 

  

(Figure 2)

(Figure 3)

My doubts are as follows: 

1. I have never used an evaluation board, the discrete resistances and capacitances shown in the circuit schematic are provided in the box, or do I have to order them separately?

2. The 'REF' point is directly connected to GND via a 0-ohm resistor. I have seen here that doing so is wrong, but the same is correct in the INA826EVM user guide (Figure 2, here). Am I doing something wrong here?

3. The load cell has a floating voltage connection, in some cases, it can also be grounded, but to obtain maximum strain, this is used. I hope the connection here is also correct?

4. There is some slight discrepancy (~20 mV) in the calculated voltage output and the simulated voltage, shown in fig. 3. Can I know the reason for the voltage drop?

Any kind of inputs would be very helpful. 

Thanks in advance, 

Sincerely, 

Sharan

  • Hi Sharan

    1. I have never used an evaluation board, the discrete resistances and capacitances shown in the circuit schematic are provided in the box, or do I have to order them separately?

    I was told that the INA826 is populated as shown in the BOM list and image below. 

    2. The 'REF' point is directly connected to GND via a 0-ohm resistor. I have seen here that doing so is wrong, but the same is correct in the INA826EVM user guide (Figure 2, here). Am I doing something wrong here?

    If Figure 6 is used, then R5 cannot be 0Ω resistor (it will short out the ref voltage from U2). If you use Figure 3 configuration, R5 should be 0Ω resistor or REF pin connected to ground. 

    3. The load cell has a floating voltage connection, in some cases, it can also be grounded, but to obtain maximum strain, this is used. I hope the connection here is also correct?

    If the strain gauge has a sensitivity of 2.0mV/V and 5 N-m is the maximum torque, you should use higher excitation voltage in order to increase the strain gauge response. 7.5Vdc* 2.0mV/V = 15mV at the INA826's differential input should be equivalent to 5 N-m in compression. (strain gauge R is 350Ω means no load, R>350Ω is tension force and R<350Ω is compression force). 

    The strain gauge should be configured as shown below. I simulated the supply voltages to +/-15Vdc. +/-7.5V will work, where the Rsg's working range is approx. +/-2.5-3Ω from 350Ω. 

    INA826 Strain Gauge 09202021.TSC

    4. There is some slight discrepancy (~20 mV) in the calculated voltage output and the simulated voltage, shown in fig. 3. Can I know the reason for the voltage drop?

    ~20mV is seen in your graph comes from the offset voltage (Vos) of INA826 part. The Vos_initial is approx. from 240uV to 850uV (RTI, refer to input) with G=1) at 25C.

    Since your configuration has Gain = 495 V/V, Vos is calculated to be approx. 20mV. 

    In order to get rid of the offset voltage, you need to calibrate the strain gauge from 0 N-m to 5N-m or user defined maximum force. For instance, if this is a balance application, zero_out the weight reading is equivalent to Tare weight function in a scale application.

    If you have additional questions, please let us know. 

    Best,

    Raymond 

  • Dear Raymond, 

    I'm elated to find such a descriptive answer and such a quick reply too! My doubts are clarified and I'll be moving forward with the hardware.  

    Many many thanks! _/\_  

    Sincerely

    Sharan

  • Hi Sharan,

    Glad that we are able to assist. 

    I am going to close this inquiry. If you have additional questions, you may still open this thread or create a new one via E2E forum. 

    Best,

    Raymond