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.

LM324: Different behaviour for two different production batches

Part Number: LM324

Hi. 

We have been using LM324N in our device since one year ago. We have never had a problem with this component, unitl now.

Our device began to fail and after several tests in several devices, we have determined that the failure occurs ONLY when we used a different batch of LM324N.

Here is attached 2 images of both different batches.

The batch that haver never had any failure is the one that is soldered in PCB. TI logo smaller.

The FAILED batch is the one that is the image with blank background. TI logo bigger.

Hope anyone help us.

  • Damian,

    There is 13 month and an assembly site difference between the two units. I believe the two units are symbolized correctly. Neither of these two differences should change device performance.

    In what way does the device perform incorrectly?
  • Hi Ron,

    Thank you for your answer.

    The failure is the following:

    - We use this LM324 in a circuit where we input a thermocouple signal and the LM324 allows its output 1 when the signal comparison is the right one. 

    - The problem is that with the FAIL LM324, no matter what the input signal is, the output1 is always ON.

    - Then we change for the RIGHT LM324 from the good batch, and in the same board, the circuit work perfect! Just changing the batch of the LM324.

    Hope this explanation was clear and wait for your feedback.

    Thank you!

  • Damian,

    Thermocouple voltages are low and input offset error voltage for LM324 can be significant compared to thermocouple input voltage.
    Have you measured the input voltages on the LM324? the difference must be greater than 7mV to ensure correct output high or low.
    Offset error changes device to device and amplifier to amplifier. However offset voltage distribution doesn't change batch to batch.

    Do all the sample from the suspect batch fail?
    How many samples were tested?
  • Damian,

    Thread marked closed due to no replies. Reply to reopen. Mark as resolved if no further action needed.