I am looking at various datasheets and was trying to understand the terminology and the implications around "input common mode range" or similarly worded terms.
For example I am looking at the TLE2072. In the absolute maximum ratings it says:
Differential input voltage range = Vcc+ to Vcc- (note: Differential voltages are at the noninverting input with respect to the inverting input).
How am I meant to interpret this ? I can see two options.
Option A:
You are free to apply any voltage to the Vin+ and Vin- as long as it is between Vcc+ and Vcc-.
Option B:
You can apply any voltage to Vin+ and Vin- as long as Vcc- <= (Vin+ - Vin-) <= Vcc+
Option A practically means do what you like with the inputs unless you have some voltages exceeding your rails somehow.
Option B is way more loose and restrictive at the same time. It implies I can apply any voltage I like to the inputs, even if I exceed the rails, as long as the difference between the two inputs is between Vcc+ and Vcc-.
As a practical example of the implication of Option B, if I had a +/- 15V rails, I would be able to apply 50V to Vin+ and 45V to Vin-, since the difference is only +5V. I would also be able to apply 10V to Vin+ and -25V to Vin-, since the difference is only -15V, within the range.
Further down the datasheet we read the "recommended ranges". There for example it says that the voltages applied at the inputs must stay within -11.9V and 15V for a +/- 15V rails. I interpet this to mean "do what you like as long as any of the two inputs do not fall outside of those two values". But in that case I could be violating the Option B above, because the differential input may then get outside the limits.
Suffice to say that on the simulator I have not managed to reproduce weird behaviours , eg phase reversals at the output when input voltages fall outside the recommended ranges. I might try it on breadboard though.
Could someone please explain this to me ?
For example I am looking at the TLE2072. In the absolute maximum ratings it says:
Differential input voltage range = Vcc+ to Vcc- (note: Differential voltages are at the noninverting input with respect to the inverting input).
How am I meant to interpret this ? I can see two options.
Option A:
You are free to apply any voltage to the Vin+ and Vin- as long as it is between Vcc+ and Vcc-.
Option B:
You can apply any voltage to Vin+ and Vin- as long as Vcc- <= (Vin+ - Vin-) <= Vcc+
Option A practically means do what you like with the inputs unless you have some voltages exceeding your rails somehow.
Option B is way more loose and restrictive at the same time. It implies I can apply any voltage I like to the inputs, even if I exceed the rails, as long as the difference between the two inputs is between Vcc+ and Vcc-.
As a practical example of the implication of Option B, if I had a +/- 15V rails, I would be able to apply 50V to Vin+ and 45V to Vin-, since the difference is only +5V. I would also be able to apply 10V to Vin+ and -25V to Vin-, since the difference is only -15V, within the range.
Further down the datasheet we read the "recommended ranges". There for example it says that the voltages applied at the inputs must stay within -11.9V and 15V for a +/- 15V rails. I interpet this to mean "do what you like as long as any of the two inputs do not fall outside of those two values". But in that case I could be violating the Option B above, because the differential input may then get outside the limits.
Suffice to say that on the simulator I have not managed to reproduce weird behaviours , eg phase reversals at the output when input voltages fall outside the recommended ranges. I might try it on breadboard though.
Could someone please explain this to me ?
There are absolute maximums and when these are exceed the IC may be damaged.
From data sheet:
absolute maximum ratings over operating free-air temperature range (unless otherwise noted)†
Supply voltage, VCC+ (see Note 1) . . . . . . . 19 V
Supply voltage, VCC− (see Note 1) . . . . . . −19 V
Differential input voltage range, VID (see Note 2) . . . . . VCC+ to VCC−
Input voltage range, VI (any input) . . . . VCC+ to VCC−
Input current, II (each input) . . . . . . . . . . . ±1 mA
Output current, IO (each output) . . . . . . . ±80 mA
Total current into VCC+ . . . . . . . . . . . . . . . . 160 mA
Total current out of VCC− . . . . . . . . . . . . . 160 mA
The other term you have seen is "common mode input voltage range". This is for normal operation.
As for "Differential input voltage range", in a normal feedback operational amplifier circuit (that does not enter into output clipping) you will not have to worry about this as the differential input is 0V.
If you were to use it as a comparator then this number is important as some op-amps have diode clamping (protection) on the inputs.
Not all op-amps exhibit phase reversal and I am not knowledgeable in simulations to say whether or not the sim would demonstrate this in an op-amp that has this characteristic.
Hope that helps.
In fact the absolute maximum ratings say :
The second line makes your option B impossible. The opamp does not have specific circuits for correctly handling voltage over the rails anyway.
So it becomes :
Option A: You are free to apply any voltage to the Vin+ and Vin- as long as it is between Vcc+ and Vcc-.
New Option B: You can apply any voltage to Vin+ and Vin- as long as Vcc- <= (Vin+ - Vin-) <= Vcc+ and both inputs are between Vcc+ and Vcc-
Option B is entirely contained in option A then.
Now, in normal use, even if the output clips or is shorted to ground, the differential voltage will not exceed option B. It could exceed it if the output is shorted to one of the supplies, or if the opamp is used as a comparator, which is a bad idea anyway.
So it doesn't matter a lot...
> Further down the datasheet we read the "recommended ranges".
> There for example it says that the voltages applied at the inputs must stay within
> -11.9V and 15V for a +/- 15V rails.
"Absolute maximum ratings" means that the device is not guaranteed to survive above those ratings.
"recommended ranges" and "operating conditions" is the range where the device will operate to specification.
Between the two, the device will survuve, but it will misbehave, you can get phase inversion, etc.
Many opamp models only include stuff like frequency response, noise, and offset, not what happens outside the normal operating range, since that would mean simulating the complete opamp circuit, with all the details, which the manufacturer is probably not going to share. So it is unlikely this will show up in simulations.
The second line makes your option B impossible. The opamp does not have specific circuits for correctly handling voltage over the rails anyway.
So it becomes :
Option A: You are free to apply any voltage to the Vin+ and Vin- as long as it is between Vcc+ and Vcc-.
New Option B: You can apply any voltage to Vin+ and Vin- as long as Vcc- <= (Vin+ - Vin-) <= Vcc+ and both inputs are between Vcc+ and Vcc-
Option B is entirely contained in option A then.
Now, in normal use, even if the output clips or is shorted to ground, the differential voltage will not exceed option B. It could exceed it if the output is shorted to one of the supplies, or if the opamp is used as a comparator, which is a bad idea anyway.
So it doesn't matter a lot...
> Further down the datasheet we read the "recommended ranges".
> There for example it says that the voltages applied at the inputs must stay within
> -11.9V and 15V for a +/- 15V rails.
"Absolute maximum ratings" means that the device is not guaranteed to survive above those ratings.
"recommended ranges" and "operating conditions" is the range where the device will operate to specification.
Between the two, the device will survuve, but it will misbehave, you can get phase inversion, etc.
Many opamp models only include stuff like frequency response, noise, and offset, not what happens outside the normal operating range, since that would mean simulating the complete opamp circuit, with all the details, which the manufacturer is probably not going to share. So it is unlikely this will show up in simulations.
Last edited:
AH OK somehow I completely missed the "Input Voltage Range" which indeed restricts the inputs to within the rails. Still that leaves me with a problem however, "New Option B" is limiting the voltage between the two inputs to half the supply. For example if Vin- is -10V and Vin+ is +10V, there we have a problem as the differential input now is 20V, which is 5V outside the limits. And these were the absolute maximum limits, which means we will likely destroy the op-amp ?
"if Vin- is -10V and Vin+ is +10V,"
This would not happen in a linear amplifier mode but would happen if you used the op-amp as a comparator...the output would be as close to the +ve rail as it could go.
I am not sure why you would want this in your audio chain.
But to your question:
If your supply was +/- 5V then you would need a new op-amp.
If your supply was +/- 15V then the output would be very close to +15V.
The op-amp would not be destroyed.
Got to go do some exercising. Bye for now.
This would not happen in a linear amplifier mode but would happen if you used the op-amp as a comparator...the output would be as close to the +ve rail as it could go.
I am not sure why you would want this in your audio chain.
But to your question:
If your supply was +/- 5V then you would need a new op-amp.
If your supply was +/- 15V then the output would be very close to +15V.
The op-amp would not be destroyed.
Got to go do some exercising. Bye for now.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.