-2

I am designing a comparator circuit based on a simple opamp (not a dedicated comparator). But, I read that using an opamp as a comnparator is foolish and undesirable. Why so ? A comparator is nothing but an opamp. WHat is the reason? Also, the reference I am giving a voltage via voltage divider and a Zener. This is shown in fig below -

Circuit

This zener will make sure that the voltage at inverting terminal is constant 3V. The other input is fed ideally 3V and when it shoots up to , say 5V, the output of the opamp goes high to 5V.

Now, in one of my posts it was stated that there should not be a resistor in parallel with zener . Why so ? If I dont step it down how can the refernce voltage work ?

Community
  • 1
Board-Man
  • 1,919
  • 1
  • 26
  • 70
  • 4
    See: [Difference between differential op amp and comparator](http://electronics.stackexchange.com/q/72967/25328) ... See also: http://circuitcellar.com/ee-tips/op-amp-versus-comparator-ee-tip-128/ – Tut Oct 15 '15 at 15:11
  • y is my doubt down voted ? – Board-Man Oct 15 '15 at 15:25
  • You've got ~25mA of diode bias current... 1 or 5 mA is more typical. (read your spec sheet!) – George Herold Oct 15 '15 at 15:35
  • 1
    I did not downvote this, but you may have unintentionally put someone off as your question may come across a little "stand-offish" when you state "Why so ? A comparator is nothing but an opamp.". Also you could improve your schematic by adding reference numbers, especially since you ask about one of the resistors. We know you are talking about the parallel resistor, but an answer may need to refer to either one. – Tut Oct 15 '15 at 15:38
  • Thank u for explaining so clearly. I will try to improve – Board-Man Oct 15 '15 at 15:39
  • @GeorgeHerold. How did you get the 25mAmps bias current value please ? – Board-Man Oct 15 '15 at 15:40
  • @VinodKaruvat 28V-3/1k = 25mA goes through the zener. What is the purpose of that 1k next to the zener ? It doesn't do anything. – efox29 Oct 15 '15 at 15:43
  • Please refrain from asking [too many questions](http://meta.stackexchange.com/questions/246328/dealing-with-bundle-omnibus-list-of-questions-question-that-consists-of-rather-d) in the same post. While asking everything you're confused about in one post is convenient for you, it diminishes the future value of the question/post, which is not how this site is intended to operate. Also, please use the search function before asking a generic question. Chances are good someone before you has had the same (generic) question. – Fizz Oct 15 '15 at 19:47
  • As for your latter question, I seem to recall @Sphero Pefhany gave you that tip in a previous question, so you could ask him for a clarification. However, I cannot seem to find that question of yours in your profile so maybe you've deleted it or I remember this wrong. (In general it's bad etiquette to delete questions with sensible answers.) – Fizz Oct 15 '15 at 19:51
  • My post was deleted by Dave Tweed . – Board-Man Oct 15 '15 at 20:03
  • If you have "read that using an opamp as a comparator is foolish" then please post a reference to where you've read that. It helps to know what your point of departure is. – Nick Alexeev Oct 19 '15 at 04:20

1 Answers1

2

A comparator is like an op-amp - but is optimised to give a digital on/off output, and the inputs are typically at widely different voltages. General op-amps are optimised to operate in the linear region where the inputs are essentially at the same voltage.

You might also want to think about adding positive feedback - to make it act as a Schmitt trigger (with hysteresis on the voltage threshold) and reduce the effect of noise at the comparison voltage.

Finally, the resistor across the zener isn't necessary - the zener will reach its operating voltage due to current flowing through the other resistor. But you might want to consider a band-gap reference instead of a zener for improved accuracy.

Icy
  • 2,253
  • 9
  • 16
  • The positive feedback when I use an Opamp, right? The comparator would be very straightforward. – Board-Man Oct 15 '15 at 15:26
  • no, you can use positive feedback with comparators too. the aim is to provide two voltage thresholds eg if you want to switch at 3V you actually switch on when the input voltage exceeds say 3.01V, but don't switch off again till the voltage falls below 2.99V - this gives a level of immunity to noise. – Icy Oct 15 '15 at 15:31