The comparator LM139 output is connected to the input by a large resistor

Thread Starter

zaizhetian

Joined Aug 30, 2017
6
Seeing others' comparator circuit, all of the comparator outputs are connected to anode on the input side through the resistance of grade M Ω. What is the purpose?

Some also are connected to anode on the input from the correct pole through resistance and diode. I have found that no one is connected to the negative side.

What is the principle?

Here is the LM139's datasheet http://www.kynix.com/Detail/861098/LM139.html.
 

ericgibbs

Joined Jan 29, 2010
10,018
hi,
You should look up hysteresis, it is added to slightly increase the comparator input voltage level, so that the comparator makes a 'firm' switch over.
Without the hysteresis, the comp output could switch rapidly high/low at the transition input voltage.
E
 

shortbus

Joined Sep 30, 2009
7,676
hi,
You should look up hysteresis, it is added to slightly increase the comparator input voltage level, so that the comparator makes a 'firm' switch over.
Without the hysteresis, the comp output could switch rapidly high/low at the transition input voltage.
E
I have a question on this Eric, I was under the assumption that the LM139/339/393 family of comparator's had 'internal' hysteresis. That this internal hysteresis was what made them different than an opamp used as a comparator.
Comments?
 

AnalogKid

Joined Aug 1, 2013
8,448
"
I was under the assumption that the LM139/339/393 family of comparator's had 'internal' hysteresis.
"

Nope. Some logic gates have hysteretic input stages, but analog comparators usually do not because it limits the usefulness of the part to only applications that need hysteresis. There are all kinds of parts and variations, but speaking in general terms...

Comparators are assumed to be used in applications that do not need a linear output. Because of this, they frequently are not internally compensated to prevent oscillation at low closed loop gains. This decreases the propagation delay through the chip. Also, they don't worry about output stage crossover distortion because the output is a digital signal. Often there is no pull up transistor, which is why the 339/393 parts heed pull up resistors on the output.

ak
 

shortbus

Joined Sep 30, 2009
7,676
Thank you both, Eric and Scott. I can't remember where I got that idea of internal hysteresis, but it was from either here or ETO in a thread. I will change my thinking on this. Again thanks.
 

wayneh

Joined Sep 9, 2010
16,390
Thank you both, Eric and Scott. I can't remember where I got that idea of internal hysteresis, but it was from either here or ETO in a thread. I will change my thinking on this. Again thanks.
I can confirm from experience that the LM339 has no useful hysteresis by itself. A swing of less than 1mV at the inputs will switch the output back and forth. That becomes most relevant when comparing voltages near ground (where there is also the offset problem). A slight rise in the ground voltage, due to a load somewhere on the circuit and an inadequate ground wire return, is plenty enough to shift the comparator state.
 

ebeowulf17

Joined Aug 12, 2014
3,274
Thank you both, Eric and Scott. I can't remember where I got that idea of internal hysteresis, but it was from either here or ETO in a thread. I will change my thinking on this. Again thanks.
Well, you're not totally crazy to have thought that. The Maxim app note that @ericgibbs attached includes this:
Hysteresis is designed into most comparators, usually with a value of 5mV to 10mV. Internal hysteresis helps the comparator avoid oscillation due to small amounts of parasitic feedback. Though sufficient to prevent the comparator from self-oscillating, such internal hysteresis can easily be swamped by any external noise of greater amplitude. In that case, you can improve performance by adding hysteresis externally.
 

ScottWang

Joined Aug 23, 2012
6,893
Well, you're not totally crazy to have thought that. The Maxim app note that @ericgibbs attached includes this:
I found out that the comparators made by Maxim as MAX9015~MAX9020 comparators have internal hysteresis and they setup to about 5~10mV, and some made by other companies, but I still can't find out the internal structures to check how they work, these comparators didn't look like the common type as many companies have the same products such as LM339 or LM393, so they should be called special comparators.

Comparator -- wiki had described some more other type of comparators has internal hysteresis.
 

Attachments

Top