• State
  • Locked Locked
  • Replies 6 replies
  • Subscribers 47 subscribers
  • Views 337 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

INA250: Instability with OPA549 with INA250 in feedback loop

Patrick Rossiter
Prodigy 40 points
Part Number: INA250
Other Parts Discussed in Thread: OPA549

Hi there,

I am trying to create a bipolar transconductance amplifier using the OPA549 as the current driver and the INA250 as shown in the circuit diagram below. The output of the OPA549 goes immediately through the INA250, which measures the current and provides a voltage signal, which is fed back to the inverting input of the OPA549. As the INA250 produces voltage centred around Vin/2, it is necessary to use an opamp to subtract the offset as shown below.

Unfortunately, when constructing this circuit I have found it to be unstable when the input voltage is negative (see oscilloscope measurements). I suspect this is the case due to the feedback opamp's bandwidth being too low relative to the OPA549, but I do not know how to confirm this before revising the circuit. For what it’s worth, I cannot replicate this behaviour in TinaTI.

Any help or guidance would be greatly appreciated. Thanks in advance.

Circuit schematic

Oscilloscope - blue is OPA549 non-inverting input, magenta is OPA549 output

  • 0
    TI__Mastermind 31915 points

    Hello Valued Engineer,

    I am looking over this and will respond shortly.

    Sincerely,

    Peter

  • 0
    TI__Mastermind 31915 points

    Hello,

    I am having trouble understanding why ripples/oscillations at VG1 (OPA549) non inverting input would indicate instability. Isn’t VG1 an independent voltage supply/waveform generator?

    Is the Opa549 output normal in this scope shot?

    Best,

    Peter

  • 0 in reply to Peter Iliya
    Prodigy 40 points

    Hi Peter, thanks for your reply. I expected the output to go to the negative rail when the input voltage was negative. Instead, it appears to flick between positive and negative rails, but mostly on the positive rail. I don't believe the input is rippling, the magenta trace is on top of the cyan one so I think it is just drawing over the input. Attached is another plot with a square wave input showing similar performance.

     

  • 0 in reply to Patrick Rossiter
    TI__Mastermind 31915 points

    I do see now, than you for the clarification Patrick.

    Are the probes/scope inputs using 50 Ohm termination? I ask because you say it’s switching between rails (.+/-15V), but scope measurements show 7V.

    Also, so ideally speaking is the input of the INA250 supposed to go below -0.3V when OPA549 input is negative? This is going to be a problem if so because -0.3 V is the devices absolute rating. The INA250 does not operate with inout voltages going below -0.3V.

    Maybe what’s happening is the OPA tries to drive a negative current with a negative output voltage. Once the INA250 input drops below -0.3V, the INA250 is trying to drive a VOUT less than 2.5V, but the linear operation stops and thus the feedback signal is stopped and OPA549 output goes back high. Once this happens, the INA starts to settle back into linear region and then sees the negative current and tries again to provide a feedback, but Vcm < -0,3 V stops linear operation and the instability repeats.

    Sincerely,

    Peter

  • 0 in reply to Peter Iliya
    Prodigy 40 points

    The supply is +/-9V in this the actual example (not +/-15V as in TinaTI), so the maximum output is only 7 V. 

    The output of the INA250 should be 2.5V when there is no current flowing. That output is then adjusted to 0V by U2. This is correct when the OPA549 is not connected, however when soldering it on to the board the issue occurs. 

    I just noticed that the INA250A4 datasheet shows a bandwidth of only 11 kHz, while the OPA549 has a bandwidth of 900 kHz. Is this likely to cause a problem having such a slow device in the feedback loop?

  • 0 in reply to Patrick Rossiter
    TI__Mastermind 31915 points

    Hey Patrick,

    Would you please confirm that the INA250 input common mode voltage (same as the OPA549!putput) is not going below -0.3V when OPA549 is trying to drive negative current?

    This is fundamental to the INA250 operation. It’s input common mode voltage can only go to -0.3V.

    The low bandwidth could be playing a role too absolutely. And to fix this you would need to slow down the whole circuit but this may not be possible for your system requirements.

    Sincerely,

    Peter