0
\$\begingroup\$

Hi I have this weird thing going on and I have no idea why.

This is my input enter image description here

And this both my input and output. The voltage is following, but still 10mV off is this normal due to being a real world application and not perfect? But none the less, where did 300mV go? enter image description here

schematic

simulate this circuit – Schematic created using CircuitLab

Is it because I am shorting my signal to "virtual Ground" ?

\$\endgroup\$
  • \$\begingroup\$ Should those crossing wires between R5 and R6 be connected? As it is, I don't see th epoint of those resistors. \$\endgroup\$ – Hearth Nov 23 '18 at 2:19
  • \$\begingroup\$ Can you explain to us what "weird thing/phenomenon" you're talking about? Looks like your circuit is following the laws of physics. In other words, can you tell us what you're expecting? \$\endgroup\$ – KingDuken Nov 23 '18 at 2:19
  • \$\begingroup\$ @Felthry Yeah you're correct just a type, fixed it. \$\endgroup\$ – Pllsz Nov 23 '18 at 2:21
  • \$\begingroup\$ @KingDuken I mean, why do you think this is normal? Should I not see 1V on the other side of my buffer? I am expecting to see 1V on the other side not 700mV. \$\endgroup\$ – Pllsz Nov 23 '18 at 2:22
  • \$\begingroup\$ @Pllsz I apologize. I didn't see the voltage reading on the oscope reading :) Did this sudden 300mV drop occur when you plugged in your oscope probes? \$\endgroup\$ – KingDuken Nov 23 '18 at 2:33
1
\$\begingroup\$

The LM833 recommended operating voltage is 10V (+/-5V) minimum and you're giving it only 7.5V.

Also your 200 ohm resistors are WAY too low, and are loading the output of the op-amp (100 ohm equivalent load). Try 4.7K or so each. That means you can also decrease C2 proportionally.

|improve this answer|||||
\$\endgroup\$
  • \$\begingroup\$ So I have tried using 10V and 12V it increased from 700mV to 810mV. Sorry update it hasn't changed anything. Still at ~700mV \$\endgroup\$ – Pllsz Nov 23 '18 at 2:52
1
\$\begingroup\$

Your supply rails are well below the opamps specified minima, but also your mid rail generator is referenced to nothing and even if it was, it would just be shorting the input signal.

Connect the junction of C3,4 to ground, and place a 10k resistor between the junction of R5,6 and the non inverting input of the opamp.

This 10k resistor sets the input impedance of the circuit, and is reasonable with the LM833 (a bipolar input opamp having fairly high bias currents), with a fet input part like your TL081 you could reasonably go much higher (100k - 1M).

Obviously this resistor forms a potential divider with the output impedance of the signal source, so if (for example) the signal generator had an output impedance of 50R then you would get a signal level drop by about 0.5%, but you would see this at the input of the opamp, if the generator was 100 ohms output you would be losing 1% and so on.

It seems likely that your signal generator has an AC coupled output (it has a capacitor in series with the output), which is why the bias arrangement works, you usually want to have a cap at the input to anything you are trying to explicity bias.

Also, make R3,4 more like 10k so the load on the first opamp is more then 100R, no general purpose opamp is happy driving a load that low, and the 081 has a particularly anemic output capability.

|improve this answer|||||
\$\endgroup\$
  • \$\begingroup\$ Yup,this did it. You got it. May I ask how did you find the value of 10K and why increase R3 and R4? \$\endgroup\$ – Pllsz Nov 23 '18 at 3:03
  • \$\begingroup\$ Sorry, I thought you meant the resistors on the split supply. Is it still normal to have 990mV from 1V still? is this due just being no op amp is perfect? \$\endgroup\$ – Pllsz Nov 23 '18 at 3:12
  • \$\begingroup\$ and I dont understand how is that 10K isolating GND from the signal, how is it still not shorting or some what shorting it and at the sametime how is the V+ pin of that buffer sees it as a reference still \$\endgroup\$ – Pllsz Nov 23 '18 at 4:57
  • \$\begingroup\$ Thats what I thought okay about the AC coupled otuput, but even if you put a cap to isolate the signal and GND would it still short?? and now if I plug in GND of the split rail I have a precision rectifier and its negative cycle gets loaded? I mean if your input is already Offset-ed to zero volts why do you need to reference V+ pin then? Why does it not work with out it \$\endgroup\$ – Pllsz Nov 23 '18 at 17:42
  • \$\begingroup\$ @Pllsz: You have accepted this answer which indicates that your problem is solved yet you have four comments with a load of questions. What's going on? Unaccept until your question is answered. \$\endgroup\$ – Transistor Nov 23 '18 at 17:51

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.