# Saturday Night Buffer Blues

Discussion in 'General Electronics Chat' started by DanRilley, Dec 11, 2010.

1. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
Hi, I am trying to make a simple Op-Amp Buffer with a TL072, but failing miserably and things are happening that I don't understand. I've attached an image of the circuit.

Basically I have an external variable resistor (big piece of carbon) that ranges between 500KOhm and 1500KOhm when agitated. I then convert this to a voltage using a divider and end up with a voltage between 3V and 6V at the Test Point pictured on the switch. This is the goal voltage.

The weird thing I don't understand is that when I connect the switch to the non-inverting input of the TL072 that point's voltage gets fixed at around 8V, even when I vary my external resistor. Also this 8V voltage is reflected at the output of the OpAmp. How do I make it so the 3V-5V I am getting at the Test Point are reflected at the output of the buffer.

I realize the TL072 needs a bipolar supply and I've tried adding a voltage divider consisting of 2 100K resistors at the non-inverting input, but that didn't seem to fix anything.

Thanks for any direction.

File size:
12.2 KB
Views:
31
2. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
Put the resistor from the + input to ground on pin 3 on the other side of the switch. You should never leave an input floating.

3. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
The switch is only to demonstrate that I am measuring two different voltages at the intersection of R1 & R3, the difference is when I connect this junction to the TL072, the voltage jumps up to 8V instead of staying at 3-6V, my question is why?

So it's not floating, I'm not using the TL072 at all in the floating configuration, I'm more looking at the left side of the circuit in that case.

Apr 5, 2008
15,806
2,389
Hello,

You can not connect the two 100K resistors at the non-inverting input.
You want to use it as input for your circuit.

The circuit as descibed is correct.
Take a look at the pages 15 and 16 of the attached PDF.

I also attached the PDf with single supply circuits.

Bertus

File size:
2 MB
Views:
19
File size:
163.1 KB
Views:
154
5. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
I see, I have actually used this Single Supply pdf for my research. If the circuit described is correct, what do you think the reason is for its failing: why is it not reflecting the 3-6V on the output?

Do I need to connect R3 to V/2 instead? Or are you implying that maybe I hooked it up wrong?

6. ### Wendy Moderator

Mar 24, 2008
20,772
2,540
A lot of op amps (yours especially) do strange things if brought to either rail, they invert the input (I think it is called phase reversal). It is a totally abnormal mode. Yes, you must use ½ the voltage.

Creating a Virtual Power Supply Ground

Apr 5, 2008
15,806
2,389
Hello,

What happens when you omit the switch?
(directly connect the 560K to the + input of the opamp).
The input impedance of the opamp should be high enough (10^12 Ohms) to have any influence on the voltage divider.

Bertus

8. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
But I'm not really bringing it to either rail am I? I'm just giving it a voltage between 3-6 volts on a 9v supply and then its giving me 8V out the end for no reason. That doesn't seem like an inversion. Sorry if I don't understand, this stuff always confuses me.

9. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
Hey Bertus, the effects I explained earlier is what happens when I omit the switch. I measure with my meter 8V on the + and 8V on the output of the TL072, I don't know where that voltage is coming from.

Apr 5, 2008
15,806
2,389
Hello,

It looks like the TL072 is defective.
Did you also try the other half of the opamp?

Bertus

11. ### DanRilley Thread Starter Senior Member

Jan 13, 2008
107
0
OK I feel like an idiot. My ground wire to the TL072 had broken off and I didn't notice it. Ahhh these saturday night buffer blues. I reconnected and now I'm seeing the values reflected on the output with a drop of about 1V. Thanks for sticking with me guys. I did learn a bit even though it was a stupid error.

Apr 5, 2008
15,806
2,389
Hello,