• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 48 subscribers
  • Views 211 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.

OPA828: High input current during power up and power down

John Chester
Prodigy 10 points
Part Number: OPA828
Other Parts Discussed in Thread: OPA827,

Tool/software:

I'm using the OPA828 as a preamp for a reproduce head in an analog magnetic tape recorder.  This is the first time I've tried using a FET-input opamp for this job, because opamps prior to the OPA827 & OPA828 had too much LF noise.  

Power supply is +/- 12 volts.   The head is directly connected to the + input of the OPA828.  When testing my prototype, I found that the head had become slightly magnetized, which is not good.  Subsequent bench testing of the OPA828 shows that + input current peaks at about 40 microamps as power supplies transition from 0.6 to 3.0 volts even if the + and - supplies remain perfectly matched.  If they are not perfectly matched peak current can be as high as 100 microamps.

My solution in the prototype was to add a power supply monitor and a relay which shorted the head when power supplies were below 90% of nominal.  This works, but it increases the cost.  

Does the OPA827 have the same problem?  Do other FET opamps?  I can test an OPA827 but I haven't done that yet.

Any suggestions?

  • 0
    Guru 303750 points

    The absolute maximum ratings forbid input voltages above V+ or below V−. Can this happen during power up?

  • 0 in reply to Clemens Ladisch
    Prodigy 10 points

    No.  The signal source connected to the + input is referenced to ground.  The - input has a feedback network with a 2k feedback resistor and a 62 ohm resistor to ground.  I have confirmed that the power supplies never reverse polarity.

    I have bench tested this circuit with an ammeter connected from the + input to ground, and if I set the power suppies anywhere between +/- 0.6 volts and +/- 3 volts I get substantial input current.  True, that's below the minimum power supply voltage of +/- 4 volts, but the input voltages do not exceed the max spec.

  • 0
    TI__Guru 66241 points

    Hi John,

    John Chester said:
    My solution in the prototype was to add a power supply monitor and a relay which shorted the head when power supplies were below 90% of nominal.
    John Chester said:
    the OPA828 shows that + input current peaks at about 40 microamps as power supplies transition from 0.6 to 3.0 volts even if the + and - supplies remain perfectly matched.

    Could you provide me the schematic of the circuit? 

    Similar to Clemens questions, are there inrush current events during the power up sequences that resulted the issue? OPA828 and OPA827 are both low noise precision op amps. Are you using switching or LDO power rails for the +/-12Vdc supply rails in the application. Please provide us these details and schematic.

    Best,

    Raymond

  • 0 in reply to John Chester
    TI__Guru 66241 points

    Hi John,

    John Chester said:
    if I set the power suppies anywhere between +/- 0.6 volts and +/- 3 volts I get substantial input current.  True, that's below the minimum power supply voltage of +/- 4 volts, but the input voltages do not exceed the max spec.

    I did not see the comments while I am replying to your query. Could you send us a schematic and I will analyze the possible root causes. 

    Could you also measure the inrush current of the +/-12Vdc rails and confirm that there is not transient events during power up? What happened if the op amp's supply rails are ramp up in a given rate from anywhere between +/- 0.6 volts and +/- 3 volt (vs. on/off power up sequences).

    Best,

    Raymond

  • 0 in reply to Raymond Zhang1
    Prodigy 10 points

    I think for the moment we can restrict this discussion to what happens during bench test of this circuit.  We don't yet need to consider all the variables introduced by the power supply in the actual device.

    There's an ammeter connected between J4 and J5.  DC power comes from a continuously adjustable bench supply with tracking outputs.  I slowly turn up the power supply voltage from zero.  When it reaches +/- 0.6 volts I start to see microamps of input current.  At +/- 1.7 volts input current on the sample I'm testing is -35 microamps.  As I continue to raise power supply voltage input current drops, reaching zero (well, below the 1 nanoamp resolution of my meter) at +/- 3 volts.

    If power supplies are not equal, input current is higher than 35 microamps when + is lower than -.  

  • 0 in reply to John Chester
    TI__Guru 66241 points

    Hi John,

    I see how you are testing the op amp scenarios. The issues are that the op amp has not reached the linear operating voltage per your testing conditions. You need to address the conditions when the op amp is reached beyond the minimum +/-4Vdc, which the linear operating conditions are established.  

    In addition, you need to perform the supply rails consistently. You need to ramp up equally on both rails. Please replace R1 to 62Ω||2kΩ, since 1Mohm will imbalance the output offset voltage. Also, increase C4 and C5 bypass to 1uF instead for your application. 

    Please let me know if the above the testing condition may help. 

    Best,

    Raymond

  • +1 in reply to Raymond Zhang1
    Prodigy 10 points

    This circuit works fine when power supply is above +/- 4 volts.  The problem for my application is the pulse of input current which occurs at lower power supply voltages.  So, this is a problem that's outside of what's on the spec sheet.  

    The value of R1 is irrelevant in my bench test, because the + input is connected to an ammeter with an internal resistance of 1.6 ohms.

    There seems to be no possible power supply sequencing in my device which will avoid magnetizing my reproduce head.  Actually, problem is worst during turn-off, because the decay rate of the supplies is a lot slower than the rise during turn-on, and thus the unwanted pulse is longer.  I think I should just go back to using a discrete FET on the input of my preamp, so I can avoid this problem.  That will be more parts, but the opamp can then be a cheaper part with higher noise.

  • 0 in reply to John Chester
    TI__Guru 66241 points

    Hi John,

    John Chester said:
    This circuit works fine when power supply is above +/- 4 volts. 

    When the supply rails are below +/-4Vdc, the op amp is operating in non-linear mode. As you indicated, there is no specification on the non-linear behavior, and we do not guarantee what the non-linear operating behavior should be. You may have to have a circuit to turn on the output when the intent operating mode is established. 

    The matching input impedance may reduce the output Vos effect, since the part has auto-compensation bias current in its analog front end (this is also applicable to linear operating mode technically). 

    Some op amp has Vout enable and disable feature, but OPA828 and OPA827 do not have it. For instance, OPA320S has the feature, which it may help in some application, but you want JFET input per the requirement. 

    If you have other questions, please let us know. 

    Best,

    Raymond