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.

LM339LV: Comparator generating voltage at the input pins

Part Number: LM339LV
Other Parts Discussed in Thread: LM339

Tool/software:

Hello everyone

I am using LM339LV quad package comparators in BMS application. The comparator is supplied from 5V and the open drain output is pulled up to 3.3V with 10K resistor. Two comparators in the quad package are being used and two are floating.

The - input of the comparator are supplied 900mV through 10K resistor and the + input is supplied 0V through 1K resistor.

The problem is that the amplifier is generating voltage on the input pins, probably by sourcing currents out of the input pins. The voltage rises from 900mV to 1.2V and from 0V to 100mV on the two pins. Appears that around 10uA current is being sourced from the input pins. What could be happening. Please help.

Thank you

  • Hi Mrinal,

    The LM339LV has failsafe CMOS inputs. There should be no current flowing into, or out of, the inputs - even with no supply.

    What are you using to measure the voltage at the pins?? DMM? Scope?

    What is the supply voltage? Can you provide a schematic?

    Is it getting hot??

    Make sure you are using the "LM339" type quad comparator pinout (all inputs at the top), and not the quad op-amp pinout (outputs on the corners). We have seen that mistake made before.

    Don't float the inputs. Since it is a R-R input, tie the negative input to V+ and the positive input to GND to force the output high. That keep the unused channels from "chattering" on noise.

  • Thank you for the reply.

    The footprint that I am using has all outputs on one side (pins 13,14,15,16). The exact part number is LM339LVRTER.

    The voltage has been measured by both DMM and oscilloscope with floating probes. The voltage is constant and not pulsating. The device is not hot. The schematic is given below

    The output of one comparator is stuck midway  at around 1.5V even though the +input terminal of the comparator is lower than the -input

  • The output of comparator 2 was not pulled up. When I pulled it up to 3.3V using 10K resistor, both comparator outputs are now at 2V. Comparator 1 that was at 1.5V has also become 2V. The voltage at the input pin has also increased from 1.2V to 1.5V, just by pulling up comparator-2 output to 3.3V

  • Hello Mrinal,

    From the schematic, I see no issues. Can we also see the Channel 2 schematic?

    Something weird is happening. The channels should not affect each other like that just by adding a pull-up. Nobody else has complained about anything like this...

    Are you sure it is soldered down correctly?? Are you sure they are 10k resistors?

    Can you send me a snapshot of your layout? Make sure pin 1 and 16 are under the corner dot and it is oriented correctly?I have seen some footprints where they number the pins clockwise (1-4 across the top). Make sure your footprint agrees with the package drawing in the datasheet (1-4 CCW down the side).

    Did you hand solder it or try re-flowing? If you get too much solder on the thermal pad, it can push the package up and cause bad connections to the pads. Try tapping or pressing it. Burnt flux around or under the device can turn resistive - clean it good.

    If the layout is correct and wired correctly - it may be damaged. That is the only way current would come OUT the inputs. Have you tried replacing it?

    What is the output driving?

  • Hi

    I found the problem. The center pad was connected to ground but the actual ground pin (pin-12) was missing connection to ground. There was a mistake in the PCB. After connecting an external wire from ground to pin-12, the device is working as expected.

    Thank you for the support

  • Hi Mrinal,

    Ah! Thought there was something wrong...

    The thermal pad is indirectly connected to GND through the die bottom, so there is series resistance. The GND pin 12 is the main GND path.

    Glad you found the problem!