• State TI Thinks Resolved
  • Locked Locked
  • Replies 4 replies
  • Answers 1 answer
  • Subscribers 49 subscribers
  • Views 3359 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.

LM324A: LM324AN vs LM324N - ESD Question

Curtis Torrey
Prodigy 25 points
Part Number: LM324A
Other Parts Discussed in Thread: LM324-N, LM324, LM2902KAV, LM2902K, LM324K, LM2902

I have an old PCB that I am updating so that it will pass the 15kV air discharge test. It is using the LM324N, and LM324AN interchangeably as voltage follower. After protecting the output lines with Zeners, everything was going great. Both -15kV and +15kV hits were dissipated correctly.

However a new batch of boards arrived and they started failing what appeared to be randomly. The difference was the LM324s.  Every board that fails has the LM324AN, and the passing ones are using the LM324N. The difference from the datasheet is below. I am wondering if the Bias Current rating may be the reason for what appears to be better ESD protection?

I am guessing that output is not able to track the change to the input during the ESD event. Which may be reverse biasing one of the transistors internally, and in the case of the LM324A it’s too much to handle. Any insight into what may be going on would be greatly appreciated!

 

Supply +24V, input signal ~6.2V

 

LM324-N

LM324A

Unit

 

TYP

MAX

TYP

MAX

 

Input Offset Voltage

2

7

2

3

mV

Input Bias Current

45

250

45

100

nA

Input Offset Current

5

50

5

30

nA
  • 0
    TI__Guru*** 140026 points

    Curtis,

    Did the new batch of boards fail after a 15kV ESD test?

    LM324N and LM324AN do not share the same silicon die. There will be subtle differences, but both are compliant to the data sheet parameters.

    LM324-N (this 'N' represents National Semiconductor, not package type) is a third silicon die and has a different data sheet.

  • 0 in reply to Ron Michallick
    Prodigy 25 points
    Hello Ronald,

    Yes, well during is more accurate. The PCB supplies a constant current to a bank of LEDs. As soon as the system is struck with the +/-15kV the LM324As fail, and the mosfet that it’s driving shuts off.

    I measured the failed LM324As and more often than not there is a short VCC to ground, but not always. Sometimes, IN- and IN+ will have been damaged and read something like 100 to 200 ohms.

    For fun I replaced the LM324As on failed boards (qty 30) with the LM324s and repeated the test. In each case it passes.

    * Using LM324N, and LM324AN. Not the LM324-N.
  • 0 in reply to Curtis Torrey
    TI__Guru*** 140026 points
    Curtis,

    Replacing 30 LM324 devices is only fun if if they are in sockets. Was the ESD pulse applied directly to a LM324 pin node or did it get there indirectly?
  • 0 in reply to Ron Michallick
    TI__Guru*** 140026 points
    Curtis,

    We also have a 2kV ESD version, LM324K, LM2902K and LM2902KAV device that have ESD added to the input pins. All other versions only have their natural protection meaning that there are no ESD specific cells in the design. There is also the fallback plan to just use LM324A which has been passing.

    In the future, try not to connect LM2902 pins to connector where human ESD can enter. If this must happen, then add series resistance in addition to capacitors and shunt protection as needed.