LM2901: LM2901 : common-mode voltage and output voltage

Hello Hyunwoo, welcome to TI e2e forums

Hyunwoo SIM said:

The test condition is that one input(IN-) exceed VCC, and other(IN+) input don't exceed VCC.

The IN+ must be (in common mode range) less than VCC-1.5V at room temperature. Then output will be low. 

Sorry, I edit this post as below because my previous explanation was not very clear.

----------------------------------------------------------------------------

In datasheet It is stated as follows:
"The upper end of the commonmode voltage range is VCC+ – 1.5V; however, one input can exceed VCC, and the comparator provides a proper output state as long as the other input remains in the common-mode range"

But when I run a simulation, I get another result.
The simulation condition is that one input(IN-) exceeds VCC - 1.5V, and other(IN+) input don't exceed VCC - 1.5V.
The output of comparator in the simulation result is the middle voltage of Vcc and ground.

In my case, VCC was 3.3V, one input(IN-) was 2.1V, and other(IN+) was 1.2V so I expected that the output of comparator in the simulation result was low(0V).
But the output of comparator in the simulatin result was 1.65V.

---------------------------------------------------------------------------

This question is about a simulation, not an actual experiment. It seems the SPICE model of comparator does not reflect the statement in the datasheet because the datasheet allows that one input can reach VCC as long as the other input remains in the common-mode voltage range.

I seems this is a bug in the SPICE model. I wonder if you can fix this.

Hyunwoo,

 The model has this text 

* Error Conditions:
* If the input goes beyond the recommended input voltage range, the output will float to mid supply
* If the supplies goes beyond the recommended supply voltage ranges, the output will float to mid supply
* The real device will NOT do this.

This is well meaning warning that there is something wrong in the simulation voltages. 

Hyunwoo SIM said:

other(IN+) was 1.2V

If this voltage is fixed at 1.2V then the application is good. In my opinion, only both out of range should do this.

Below is a TI-tina classic model that doesn't try to help you. 

LM339,LM2901 Classic.TSC

Here is model in text, very limited.

* LM339 VOLTAGE COMPARATOR "MACROMODEL" SUBCIRCUIT
* CREATED USING PARTS VERSION 4.03 ON 03/07/90 AT 14:17
* REV (N/A)
* CONNECTIONS: NON-INVERTING INPUT
* | INVERTING INPUT
* | | POSITIVE POWER SUPPLY
* | | | NEGATIVE POWER SUPPLY
* | | | | OPEN COLLECTOR OUTPUT
* | | | | |
.SUBCKT LM339 1 2 3 4 5
*
F1 9 3 V1 1
IEE 3 7 DC 100.0E-6
VI1 21 1 DC .75
VI2 22 2 DC .75
Q1 9 21 7 QIN
Q2 8 22 7 QIN
Q3 9 8 4 QMO
Q4 8 8 4 QMI
.MODEL QIN PNP(IS=800.0E-18 BF=2.000E3)
.MODEL QMI NPN(IS=800.0E-18 BF=1002)
.MODEL QMO NPN(IS=800.0E-18 BF=1000 CJC=1E-15 TR=807.4E-9)
E1 10 4 9 4 1
V1 10 11 DC 0
Q5 5 11 4 QOC
.MODEL QOC NPN(IS=800.0E-18 BF=20.29E3 CJC=1E-15 TF=942.6E-12 TR=543.8E-9)
DP 4 3 DX
RP 3 4 46.3E3
.MODEL DX D(IS=800.0E-18)
*
.ENDS

Hello Hynwoo,

The model purposely does not model the "feature" of having one input within the range. This is because it is technically out of datasheet specification (VCC-2V) as applied to either input pin.

The "one input in 'feature'" is not a guaranteed parameter. It is classic expected behavior, but not guaranteed behavior. We do not want to encourage reliance on unspecified behavior, so we took the conservative route by flagging the error in the model. 

Please see section 6.7 (and the warning) of the LM339 family appnote:

Application Design Guidelines for LM339, LM393, TL331 Family Comparators

If input range is an issue, please consider using the LM339LV, which has Rail to Rail inputs and will gladly accept the above input voltages.