Need help interpretting OpAmp spec, supply voltage

[DragonPetter]

In the datasheet it says the voltage max is +/-18V.
Does this mean that the maximum on Vs+ is 18V, or that the maximum voltage between Vs+ and Vs- can be 18V, or 36V?

I want to power it with Vs+ at 36V and Vs- at 0 volts.The datasheet is here: http://focus.ti.com/lit/ds/symlink/opa627.pdf

 

[berkeman]

In the datasheet it says the voltage max is +/-18V.
Does this mean that the maximum on Vs+ is 18V, or that the maximum voltage between Vs+ and Vs- can be 18V, or 36V?

I want to power it with Vs+ at 36V and Vs- at 0 volts.


The datasheet is here: http://focus.ti.com/lit/ds/symlink/opa627.pdf

Yes, it can be powered with a +18V and -18V supply, or your 36V single-ended power supply. How are you going to make your 18V mid-rail reference voltage?

 

[DragonPetter]

Yes, it can be powered with a +18V and -18V supply, or your 36V single-ended power supply. How are you going to make your 18V mid-rail reference voltage?

What do you mean by mid-rail reference voltage? I'm a little confused and worried now.

I attached a drawing of what I plan to do.

I want to measure the voltage on a motor with an ADC in a microcontroller. The motor voltage ranges from 12V to 36V, and the ADC range is 3.3V to 0V. So I need to voltage divide the motor voltage before it goes to the ADC, but I only care about measuring 12V and up, so I use a zener to use my gain only from 10V to 36V to increase my resolution. I will use a more accurate zener than bzx84C family since the opamp currents might move around the zener knee, but I use it here just as the idea.

So I am hoping: Vout = 0.0091(Vmotor-10V)

 

[DragonPetter]

Also, I'm changing the 10k and 910 to 100k and 12.5k so that I don't unbias the zener and give the proper gain.

 

[berkeman]

What do you mean by mid-rail reference voltage? I'm a little confused and worried now.

I attached a drawing of what I plan to do.

I want to measure the voltage on a motor with an ADC in a microcontroller. The motor voltage ranges from 12V to 36V, and the ADC range is 3.3V to 0V. So I need to voltage divide the motor voltage before it goes to the ADC, but I only care about measuring 12V and up, so I use a zener to use my gain only from 10V to 36V to increase my resolution. I will use a more accurate zener than bzx84C family since the opamp currents might move around the zener knee, but I use it here just as the idea.

So I am hoping: Vout = 0.0091(Vmotor-10V)

Won't work. That opamp (as with many opamps) is designed to work with split supplies (or a single supply where the inputs and outputs are running mid-rail somehow, like with a mid-supply reference voltage).

Take a look at the input and output voltage specs in the table in the datasheet. With a +/-15V supply, the input and outputs can go to about +/-11V. That means that you need to keep the inputs and outputs at least 4V or so away from the rails. You cannot run this part (or many opamps) off of a single supply and use ground as a signal reference as shown in your schematic.

 

[DragonPetter]

Won't work. That opamp (as with many opamps) is designed to work with split supplies (or a single supply where the inputs and outputs are running mid-rail somehow, like with a mid-supply reference voltage).

Take a look at the input and output voltage specs in the table in the datasheet. With a +/-15V supply, the input and outputs can go to about +/-11V. That means that you need to keep the inputs and outputs at least 4V or so away from the rails. You cannot run this part (or many opamps) off of a single supply and use ground as a signal reference as shown in your schematic.

Ok, thank you very much Berkeman, Iæm glad I asked. Back to the drawing boards for me, but I will just try to refine this attempt. I'll show my new idea after some more work.