1
\$\begingroup\$

I need help explaining why I am getting such a large DC offset across the output for my op amp.

I am using an OPA548 to amplify an AC signal and then drive a solenoid shaker. I have the circuit designed to have a gain of 10 and I am not current limiting the op amp (the OPA548 has that capability). Before I input any signal into the op amp, I connected the output, with respect to ground on the circuit, to an oscilloscope and I saw that there was a DC offset around 13.3V.

To clarify more, I am powering the op amp with +-15V and amplifying a 1V pk-pk input to 10V pk-pk. I then adjusted the voltage that I am powering the op amp with (+-15V) and I saw that the offset is completely dependent upon the +15V. Why is that?

I understand that op amps can have DC offset because of the internal circuitry but I never have had it where the DC offset was dependent upon the input voltage to the op amp. Then, when I applied an input signal and then looked at the output voltage signal on an oscilloscope, the DC offset seemed to "go away" because the output signal was oscillating around 0V. I don't understand why that is. The signal still looked clean at that point.

Lastly, I connected the output across the load (solenoid shaker). The picture below is what I saw on the oscilloscope. My guess as to why that is is as follows: The DC offset noticed before still must be having an affect. When I connected the shaker across the output before applying a signal (the time when I noticed the DC offset) I could feel that the shaker was energized. It felt like the weight inside of it was off to one side and operating at a high frequency. My reasoning as to why the oscilloscope graph looked the way it does is because the DC offset must be holding the weight off to one side before enough negative voltage can override and then it goes back to where it was before.

I would like your thoughts on this and potentially how I could fix this issue. I hope I was detailed enough. Please ask for more details if you think that would help. I attached pictures of both the solenoid shaker and the oscilloscope graph that I am trying to interpret. Thanks in advance for your help

Oscilloscope graph when shaker was connected across output with signal being applied Solenoid Shaker

\$\endgroup\$

closed as unclear what you're asking by Matt Young, Daniel Grillo, Keelan, Scott Seidman, Kaz Mar 24 '15 at 14:34

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ The question got lost in the wall of text. \$\endgroup\$ – Ignacio Vazquez-Abrams Mar 17 '15 at 17:47
  • \$\begingroup\$ Perhaps you have provided - unintentionally - positive feedback instead of negative feedback? \$\endgroup\$ – LvW Mar 17 '15 at 18:14
  • 2
    \$\begingroup\$ How about a circuit diagram. Looks like HF oscillations on the 'scope trace... you do have bypass caps in place, right? How does it look when you are just driving a resistive load? \$\endgroup\$ – George Herold Mar 17 '15 at 18:42
  • \$\begingroup\$ Please provide schematic diagram. We can't offer an opinion without seeing what you are trying to do. \$\endgroup\$ – Dwayne Reid Mar 19 '15 at 11:27
3
\$\begingroup\$

I'm assuming you're using a non-inverting configuration.

First, your DC offset. When you looked at the output, you had no input at all - not even a resistor to ground. So the input floated and the output went to the rail. Try connecting your input to ground with, let's say, a 10k resistor. If, for some reason, you are using an inverting configuration, ground the input. If you stll seem to have a large offset, particularly one which seems unsteady, look at it with the scope and make sure the output is not oscillating.

The reason the offset is dependent on the value is that the op amp is trying its very best to produce an output at V+. If you raise V+, the output follows it.

Next, oscillation. Take a look at the data sheet http://www.ti.com/lit/ds/sbos070b/sbos070b.pdf, Figure 1. Are you doing ALL of this? And do you have a good ground plane to connect those capacitors to? Those capacitors are there for a reason. Plus, you need a solid ground, with the power supply ground connected to the motor return right on the board close to the op amp. If you run separate leads back to the power supply you may well be forming a ground loop.

Your scope trace is pretty classic for this sort of op amp. The positive output transistors, being NPN, are faster than the negative PNPs, so a setup which is marginally stable for negative swings is unstable for positive swings.

\$\endgroup\$
  • \$\begingroup\$ Thanks for your comments. To answer your questions, I did connect the input to ground and my offset went to 0 so that explains the DC offset. Also, I did look at the data sheet and no I do not have those capacitors in there. I am going to use them next and I will post the results. In the meantime, could you explain to me what those capacitors do? Thanks again. \$\endgroup\$ – Max Moeller Mar 17 '15 at 22:45
  • \$\begingroup\$ Those are called decoupling capacitors. They "decouple" the output load from the power supply voltage, especially for high frequencies (like where your output is oscillating). Without them, a large load pulls down the supply voltage. This causes the amp to try to draw less current. So the supply voltage rises, the load current pulls it down, and around it goes. The larger caps work better at lower frequencies, and the smaller at higher frequencies. And ground quality matters. Any more detailed than that, and you need to start learning more electronics. \$\endgroup\$ – WhatRoughBeast Mar 17 '15 at 23:12
  • \$\begingroup\$ I decoupled the supplies and that fixed the problem. Thanks for your help! \$\endgroup\$ – Max Moeller Mar 18 '15 at 17:05
  • \$\begingroup\$ Glad to help. (Shameless plug follows:) Don't forget to select my answer. \$\endgroup\$ – WhatRoughBeast Mar 18 '15 at 17:12

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