• State Resolved
  • Locked Locked
  • Replies 2 replies
  • Subscribers 49 subscribers
  • Views 592 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.

Question on LM2903

Eva Pimenta
Prodigy 10 points
Other Parts Discussed in Thread: LM2903

Hi ,

My first post on a TI forum!  Woot! Woot!

I am working with the LM2903 as a comparator in a discrete circuit, to perform a relatively simple operation, a simple circuit:

- It senses the voltage on the line on the (+) input, and compares it to a steady voltage on the (-) input (the comparing voltage), with NO feedback, and merely switches the output to be in 1 of 2 states :    either Hi-Z output state or to GND output state. The output pin of the comparator has a pull-up resistor to some voltage, V_PULL_UP.

I was testing this in the lab, applying a variable DC source on the (+) line, and changing it to just confirm the two output possibilities. There is a transition region between the two outputs, where the output voltage is between GND and the V_PULL_UP votlage.

The range of transition voltages is very small, a fraction of a voltage.  I was commenting, that this seemed normal in my experience.  However, there was another engineer present, who was of the opinion, that was out of normal behavior.

I tried to take a look at the datasheet for the LM2903, but could not quickly located a metric to describe this transition range. 

Can anyone comment on this? Any ideas?  Thanks for your time!

Regards,

Evie

I think this is normal. The transition region is very small, a fraction of a voltage. However, there was another engineer, that thought this was not a good characteristic.

  • 0
    TI__Mastermind 49966 points

    HI Eva,

    Yes. This is normal behavior. The output still only has two states, but it is rapidly switching between the two states when the input voltages are very close to each other. This can look like a random voltage when measured with a DMM. The comparator is randomly triggering on the few microvolts of noise. This is very common when the signals are moving slowly.

    Please see the following TI design article:

     Comparator with Hysteresis Reference Design

    Regards,

  • 0
    Intellectual 480 points

    Valuable lesson, don't use a meter, use a scope. The meter will average the results, typically over 100mS, so noise will produce something bigger than zero and less than the supply. The scope will show the noise. (Which is why you frequently add hysteresis or use a Schmidt trigger input device.).