Op-amp part replacement comparison

[FreshmanNewbie]

I have a question.

Can I use this part (https://ww1.microchip.com/downloads.../MIC6270-IttyBitty-Comparator-DS20006294A.pdf) in place of this part (https://www.ti.com/lit/ds/symlink/l...44036&ref_url=https%3A%2F%2Fwww.google.com%2F).

The application is for ZCD where the input to the op-amp will be coming from an opto-isolator and given to the non-inverting terminal.
U21 in the attached image.

 

[KlausST]

Hi,

The current schematic is for analog output. Thus they use an amplifier / OPAMP.

If you want a digital output ... you first have to modify the schematic ... then you may use a comparator.

Generally:
comparators and OPAMPS are different devices, made for different operating modes.
* a comparator never has negative feedback. But it may have positive feedback (to generate a hysteresis).
* an OPAMP never uses positive feedback. (It would cause the circuit to act as a comparator with hysteresis. But then use a comparator IC, since many OPAMPS are not designed to operate as a comparator). A OPAMP amplifier circuit always uses negative feedback.

Klaus

 

[danadakk]

* an OPAMP never uses positive feedback. (It would cause the circuit to act as a comparator with hysteresis. But then use a comparator IC, since many OPAMPS are not designed to operate as a comparator). A OPAMP amplifier circuit always uses negative feedback.

Its more complicated than the above :

Lecture 21: Op Amps Positive Feedback | Circuits and Electronics | Electrical Engineering and Computer Science | MIT OpenCourseWare

MIT OpenCourseWare is a web based publication of virtually all MIT course content. OCW is open and available to the world and is a permanent MIT activity

ocw.mit.edu

Using OpAmps as comparators (sometimes advantageous) - https://www.analog.com/media/en/training-seminars/tutorials/MT-084.pdf

 

[KlausST]

Its more complicated than the above :

What exactly does the video show ... that is "more complicated" than what I have written?

Using OpAmps as comparators (sometimes advantageous)

Can you please point out where exactly the document refers to something "advantageous"?

Klaus

 

[danadakk]

More alluding to one could envision considerations. Maybe to help more
completely this discusses both advantages and disadvantages (like DC and
Vos inaccuracies in most comparators) -

https://www.analog.com/media/en/training-seminars/tutorials/MT-084.pdf

MT084 ap note did allude to using leftover OpAMP as a comparator for costs savings.

One can also think in terms of low noise OpAmps having an advantage over comparators.
And of course one can get power OpAmps which would make a good power comparator.
I googled power comparators to see if there are any of significance, biggest I 'found was
360 mA output current. OpAmps I found > 10A.

The video establishes one can use +fdbk as long as net is negative, linear operation (excluding
some osc topologies which utilize + fdbk).

 

[KlausST]

Hi,

The OP wants do discuss the use of a COMPARATOR instead of an OPAMP in a cricuit with 100% negative feedback.
So the use of an OPAMP with positive feedback is not the OP´s topic at all.

Sadly it was me who brought in the phrase

an OPAMP never uses positive feedback

... in the section with the headline "Generally" .. where I just wanted to give a brief information about the differences between OPAMP and COMPARATOR circuits.

*****
OFF TOPIC from here
==================

MT-084:

I neither can find the words "advantage", nor "benefit", nor "leftover", nor "cost", nor "offset", nor "VOS" ... what do I need to search for?
Maybe my PDF search function does not work.

The document starts with:

Even though op amps and comparators may seem interchangeable at first glance there are some
important differences. Comparators are designed to work open-loop, they are designed to drive
logic from their outputs, and they are designed to work at high speed with minimal instability.
Op amps are not designed for use as comparator...

and ends with:

SUMMARY
Operational amplifiers are not designed to be used as comparators, so this tutorial has been,
intentionally, a little discouraging.

and in the following:

nevertheless there are some applications where the use of an
op amp as a comparator is a correct engineering decision—what is important is to make it a
considered decision, and ensure that the op amp chosen will perform as expected. To do this it is
necessary to read the data sheet carefully, to consider the effects of non-ideal op amp
performance, and to calculate the effects of op amp parameters on the application. Since the op
amp is being used in a non-standard manner some experiment may also be necessary–since the
amplifier used for the experiment will not necessarily be typical the results of experiments
should always be interpreted somewhat pessimistically.

So is the document meant for newbies ... or rather for experienced designers?

One can also think in terms of low noise OpAmps having an advantage over comparators.

The output of a comparator .. as well as of an OPAMP with positive feedback is digital HIGH/LOW ... how much impact has the input noise to the output?
How can one measure it?
I do understand the noise measurement of an OPAMP ... in it´s "usual" operation mode as an amplifier. But have no idea how it can be measured on a comparator with hysteresis.
Comparator datasheets don´t talk much about noise ... but due to their lower open loop gain (compared to OPAMPs) .. doesn´t the input referred noise has less impact on the output than at an OPAMP?

And of course one can get power OpAmps which would make a good power comparator.
I googled power comparators to see if there are any of significance, biggest I 'found was
360 mA output current. OpAmps I found > 10A.

I don´t understand how is this related to
* the OP´s question
* the documents you refer to

Klaus

 

[danadakk]

MT-084:

I neither can find the words "advantage", nor "benefit", nor "leftover", nor "cost", nor "offset", nor "VOS" ... what do I need to search for?
Maybe my PDF search function does not work.

Yes it appears they did not use your words, or as you pointed out your self limiting search for a term,
and not reading the document, well meaning I am sure, but crippling. Did you do the same for the
second doc -

The output of a comparator .. as well as of an OPAMP with positive feedback is digital HIGH/LOW ... how much impact has the input noise to the output? How can one measure it?

Histogram over time. Use Vref with low noise, compare with noise generator with offset, and create histogram.
Or bias comparator into linear region to look at self generated noise spectrum....

Can you please point out where exactly the document refers to something "advantageous"?

Above not OP related, just adding to your “advantageous” quest.



Knight

 

[KlausST]

Yes it appears they did not use your words, or as you pointed out your self limiting search for a term,
and not reading the document, well meaning I am sure, but crippling. Did you do the same for the
second doc -

Sorry, it was not "my words" ... it was the words you gave. And since I did not find the information .. I even expanded the search with similar words.

Again you show an aggressive attitude when I ask simple questions about informations you gave. Why?
I took your informations seriously, did read through the documents and your posts, asked for details, asked for help.
I asked for "exact locations" in the document where the "advantageous" use of an OPAMP is written. (A simple answer could be: "Page#6, bottom")
I took effort in finding the location on my own. I did not find them ... so I asked for help.
What´s wrong with asking for help?

Still I can´t find the part where it gives the "several reasons to use OPAMPs as comparators".
Call me dumb, call the PDF search dumb ... which page is it in? Is this asking too much? Without getting back a disrespectful attitude...

I don´t think that my posts were disrespectful.
Mainly I asked for your assitance to clarify things you wrote and how they fit to the informations in the linked documents.

Which "second doc" are you referring to?
--> You twice gave a link to the same MT-084 doc. And the other link guides me to a 50 minutes long video
In which post is the link? Or do I have to follow links within the linked pages? Is it asking too much.. for a smiple clear answer? ... without attitude?

****

Histogram over time. Use Vref with low noise, compare with noise generator with offset, and create histogram.
Or bias comparator into linear region to look at self generated noise spectrum....

I understad this setup... (hopefully)
But how can the use of an external noise source .. find out the internal noise of the OPAMP_comparator?

And... as far as I understand .. you can´t get an OPAMP_with_positive_feedback .... into linear region. Please explain how this can be done. (without being upset and unfair)
This is as serious meant question ... for me and others to learn new things.
I understand that for you - who know how these things work - it may be an annoying and stupid question.

Klaus

 

[danadakk]

My error, second doc attached.

And... as far as I understand .. you can´t get an OPAMP_with_positive_feedback .... into linear region. Please explain how this can be done. (without being upset and unfair)

I posted MIT video with the following comment :

I had no idea this would cause you to conclude I am aggressive, that I am adverse to a forum discussing
engineering and focused on delivering ACCURATE info. Occasionally I get corrected, I consider that a
best in class process. But all these psychological conclusions of my intent and interpretation not really
useful here. I will go out on a limb and post you are not qualified in that department. Correct me if I am
wrong about this.....

The MIT video seems to me to state that NET negative feedback is desired, eg. that one can have both
positive and negative feedback and have a linear result. What "triggered" me was your statement :

* an OPAMP never uses positive feedback. (It would cause the circuit to act as a comparator with hysteresis. But then use a comparator IC, since many OPAMPS are not designed to operate as a comparator). A OPAMP amplifier circuit always uses negative feedback.

Highlight an OPAMP never uses positive feedback.

I understand that for you - who know how these things work - it may be an annoying and stupid question.

More psychological

The rest of your non OP or topic discourse I will leave for .......

 

[KlausST]

Your attitude is outstanding!

You miss to post the link .. and then you accuse people for not reading this document!

Highlight an OPAMP never uses positive feedback.

... and you intentionally ignored what I wrote thereafter.
"It would cause the circuit to act as a comparator with hysteresis...."
And that´s exactly what the MIT video shows.
In the first quarter of the video they said "it makes bang bang to either supply rail ... later we will try to make some useful funtion"
And the whole last thrid they disccussed a "a comparator with hysteresis...." (Compare this with what I have written)

Klaus

 

[danadakk]

Your attitude is outstanding!

You miss to post the link .. and then you accuse people for not reading this document!


... and you intentionally ignored what I wrote thereafter.
"It would cause the circuit to act as a comparator with hysteresis...."
And that´s exactly what the MIT video shows.
In the first quarter of the video they said "it makes bang bang to either supply rail ... later we will try to make some useful funtion"
And the whole last thrid they disccussed a "a comparator with hysteresis...." (Compare this with what I have written)

Klaus

The video shows that NET negative feedback, eg (POSfdbk - NEGfdbk) results in stable final value if that
difference is <=0.

Appeal to others on forum group, watch MIT video and confirm/deny my observation is in error.

 

[KlausST]

The video shows that NET negative feedback, eg (POSfdbk - NEGfdbk) results in stable final value if that
difference is <=0.

if:
(POSfdbk - NEGfdbk) <=0

simply means .. the negative feedback dominates....

--> I did never disagree. (You are free to show the people WHERE I disagreed to this. I bet you can´t ... because I did not)

******
And:

results in stable final value

Is an OPAMP with positive feedback ... I.e. a comparator with hysteresis considered "unstable"?

--> I did never disagree to that. (You are free to show the people WHERE I disagreed to this. I bet you can´t ... because I did not)

Klaus

 

[D.A.(Tony)Stewart]

I see no need for all these filters, an Op Amp or a comparator unless you state some useful output specifications like PW50 = 1 ms max, centred on transition, active low. My design may be preferred for glitch filtering, resistor stress reduction, and simplicity. However, you will get better advice with design specs like purpose and PW.

You can reduce PW50 to 1us if it was necessary with an XOR gate with RC delay as a 1-shot comparator on a sine wave.

--- Updated Sep 27, 2024 ---

I intended the 5.6 uF cap to be 1 uF ceramic and the design here works from 0.1 uF to 1mF+

There is no need to choose Line, Neutral with this differential design and will work on 120Vrms 60Hz..

 

[danadakk]

Your attitude is outstanding!

You miss to post the link .. and then you accuse people for not reading this document!

My statement you had not read the document was correct !. Following this I corrected my error, post
#9, by stating I made an error, and attached the doc. Then comes your Psychological "attitude" diagnosis.
Wow .....double wow.....

Lets do a reset, you state the following :

Generally:
comparators and OPAMPS are different devices, made for different operating modes.
* a comparator never has negative feedback. But it may have positive feedback (to generate a hysteresis).
* an OPAMP never uses positive feedback. (It would cause the circuit to act as a comparator with hysteresis. But then use a comparator IC, since many OPAMPS are not designed to operate as a comparator). A OPAMP amplifier circuit always uses negative feedback.

From that I conclude :

1) I positive feedback present the only destiny for OpAmp circuit is comparator action. MIT
video clearly states it as NET feedback determines behavior. And a sim I did shows as well,
although I still have questions about the sim. And I want to investigate effect on amp Z's
with both + and - fdbk present.

2) "A OPAMP amplifier circuit always uses negative feedback." Thats pontificating based on
you not knowing the application of every design in the local Galaxy group we are in. Neither
do I.

Then you went on to challenge use of OpAmp as a comparator has no redeeming values.
The first ap note alludes to this, the second in more detail. And my extraction and post of
advantages from the second ap note. Which you conveniently ignored.

Lastly you regurgitate several times the MIT video only proves positive fdbk results in comparator
behavior, and prof went out of his way to show the presence of positive fdbk NOT the only criteria
for comparator action. It has to be NET positive, eg. implying both neg fdbk and pos fdbk must
be taken under consideration.

Time to say goodbye, I am leaving the merry-go-round.

Knight

 

[KlausST]

Hi,

if you want to discuss

every design in the local Galaxy

... then maybe start your own youtube channel, TV show or what you like.
That´s exatly the problem, that you can´t focus on the OP´s "simple" problem but want to perform your own show, by discussing rare and exceptional designs.

But here
--> The OP will not gain much from this.
--> the OP also will not gain much from your MIT video.

The OP wants an answer to his question .. and not a discussion about rocket science.
He needs a guiding hand for his very problem.
I stayed at basic informations ... adressed to the OP, adjusted to his "Newbie" level.

Then you went on to challenge use of OpAmp as a comparator has no redeeming values.

No, I did not!
I gave the OP the advice not to use an OPAMP as a comparator ..... and you made your own challenge from this.
Your challenge was to make rocketscience ... from a simple OP´s question.
If your ego needs a "Galaxy discussion", then - if you think it makes sense - do this with the OP. It´s his thread, not mine.

But I stand to my statement, adressing the OP (and explicitely not you): Don´t use an OPAMP with positive feedback, don´t use an OPAMP as comparator.

Which you conveniently ignored.

Every one can see that you did not post the link.
But yes .. I´m not willing to discuss about your "Galaxy circuits". Do this challenge with someone else!

Lastly you regurgitate..

everyone can read this thread...

Time to say goodbye, I am leaving the merry-go-round.

Yes, your own built one.. in your own Galaxy

Klaus

 

[D.A.(Tony)Stewart]

I think our expertise is great but limited by experiences not had. I appreciate all inputs but we have lost feedback from the newbie who is learning about conflicts in opinions. I suggest we close this thread unless newbie says something to acknowledge what he has learnt from this. I suggest 12h from now to remind him we are not like other professionals with great wisdom and do this all for pro bono. So show some respect and participate.

My view does not differ from Dana’s on Op Amp diversity and advantages as a comparator but I lean towards using the discrete differential Vbe design I presented for simplicity. But for speed/accuracy the XOR differential delay is the ultimate from my 1st creation of this circuit in 1976 for a 8 ch 1.5kW chaser using Triacs for a musician buddy’s band. This minimized the flood lamp splash surge power 10x rated for incandescent as long as the rep rate did NOTfunction at a speed where the avg power far exceeded the full power of the lamp and Triac. Poof.

--- Updated Sep 28, 2024 ---

Those who have used variable hysteresis and Linear Sallen & Keys filters might disagree with saying Op Amps never use positive feedback.

Op Amps with limited GBW as high gain comparators have the advantage of ignoring 1~10us line transients for a ZCS function .

It is more effective to use the optoisolator as a common emitter than an emitter follower as the voltage gain is more desirable for a pulse amplifier.

My design presented was a more common ZCS implementation with a slight variation with the added NPN for greater tolerance to cheap low Beta (<1) Optos so it is far less less sensitive to component values, temperature effects and allows high input R’s to create a Norton low current source.

I do not expect a newbie to understand my design unless they focus on the signals for each part shown. Any expert ought to understand it readily and make a critical evaluation.