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.

OPA561 / the frequency specification is varied by RCL

Guru 20090 points
Other Parts Discussed in Thread: OPA561, TINA-TI

Hello,

If we change the RCL value, is the AC characteristics (GBW or Phase margine etc..) varied?

Our customer faced the output oscillation when they change the RCL value of OPA561.
Please see the following as the test conditions. I'm checking the other informtion to customer but I couldn't get.

Case 1 : RCL=39kohm, any Iout, no oscillation
Case 2 : RCL=4.3kohm, Iout= ~ 10mA , no oscillation
Case 3 : RCL=4.3kohm, Iout = 100mA~ , 10MHz, 200mVpp oscillation

Best Regards,
Ryuji Asaka

  • Asaka-san,

    Something seems odd about the output condition applied to the OPA561. If case 2 has a 4.3 k-ohm load and the output current is 10 mA, the peak output voltage would have to be 43 V. Similarly for case 3 if the load again is 4.3 k-ohm, but the output current is 100 mA, then the output voltage would be 430 Vpk. Certainly these voltage levels are well beyond the capability of the the OPA561, which has a maximum supply voltage of 16 V. If indeed these output current levels are being measured, then there must be an additional load in parallel with the 4.3 k-ohm resistor that is drawing current. There may be a capacitive load on the output that is acting as an ac load.

    Please provide the OPA561 application circuit schematic showing supply, input signal and output levels. DSO images would be helpful.

    Regards, Thomas

    PA - Linear Applicaitons Engineering

     

  • Hello Thomas san,

    RCL meaning is current limit setting register.
    Since the schematics is our customer confidential, I couldn't get yet.

    Please see the following the information.

    V+ = 5V, V- = -5V
    Vin= -0.5V ~ +0.5V , DC input only
    Av=3, (Rf=2kohm, Rs=1kohm)
    Iout= 0mA~ +/-150mA

    OPA561_20150406.pdf

    I simulated the OPA561 on the some RCL value with above conditions by TINA-TI.
    Please see the attached file.
    As a result , the some analysis result (AC transfer characteristics, Noise, transient )are same between some RCL values.

    Is not affect the characteristics other than the current limit value when we change the RCL at the Ilim pinf?

    Best Regards,
    Ryuji Asaka

  • Asaka-san,

    Thank you for the clarification about RCL. The current limit circuit is local to the output stage and I would not expect the value of RCL to affect the amplifier stability. Since the oscillation comes and goes with the particular RCL value it does appear to be having some effect.

    The low-level, 10 MHz oscillation you describe has characteristics more like one associated with a loop stability issue i.e. too much delay around the feedback loop. Often, that is the result of a large capacitive load on the output, too much capacitance placed on the amplifier summing node, or poor layout and decoupling practices. Without the customer's actual circuit and interconnection information we don't have any clues. Find out if TI has an NDA with the customer. If we do, then they should be able to share their confidential schematic with us. We would do that off the E2E forum to limit access.

    There is a great deal of op-amp and stability information available from the new, TI Precision Labs site. Here is a link to the op-amp stability sessions:

    http://www.ti.com/lsds/ti/amplifiers-linear/precision-amplifier-precision-labs.page?DCMP=tipl&HQS=hpa-pa-opamp-tipl-pp-tr-en

    This may be helpful to you, or the customer, to help better understand the amplifier stability subject.

    Your TINA simulation look good and the results are believable for a correctly operating OPA561.

    One last thing, make sure is that the OPA561 PowerPAD is connected to V-. If that connection is not made it can lead to some unexpected amplifier behaviors.

    Regards, Thomas

    PA - Linear Applications Engineering

  • Hello Thomas san,

    Thank you very much for the reply.
    I will check the customer actual schematics and Power PAD connection.
    And I will inform that the RCL does not affect to the stability of amplifier.
    I'm checking the precision labs video for my studying. Thank you.

    Best Regards,
    Ryuji Asaka
  • Hello Thomas san,

    The customer didn't connected the Power Pad to V-.
    I think that the cause of the ocillation may be disconnection of these.

    However, the customer have the other question.
    In the datasheet P.10 Figure 6, the IC(sillicon) is not electrical connected to Exposed pad.
    Could you please let me know the mechanism of oscillation when we the PowerPAD is not connected to V-?

    Best Regards,
    Ryuji Asaka
  • Hello Aska-san,

    OPA561 data sheet Fig. 6 shows a cross-section of the amplifier's physical construction. What the cross-section doesn't indicate is the epoxy adhesive, that bonds the amplifier IC die to the leadframe die pad (PowerPAD), is conductive. The epoxy has both high electrical and thermal conductivity. Therefore, when V- is applied to the PowerPAD the back of the die is biased to that voltage as is required.

    I do not know for certain if failing to apply V- to the PowerPAD is leading to the oscillation, but there is a possibility that it could be the reason for it. If the back of the die is not biased at V- the parasitic capacitances of the transistor structures can be different than what they are suppossed to be.

    Recently, I wrote a blog about the reasons and importance of connecting the PowerPAD to the correct, designated voltage. I believe it will provide you a more indepth understanding of the requirement:

    https://e2e.ti.com/blogs_/b/precisionhub/archive/2014/11/06/why-correct-voltage-matters-for-op-amp-powerpads

    Regards, Thomas

    PA - Linear Applications Engineering

  • Hello Thomas san,

    The OPA561 osillation was improved when the customer connected the Power Pad to GND.
    Thank you for your support. Sorry, I got the additional question from the customer.

    Could you please let me know the solder coverage of OPA561 Power Pad?
    Is this about 60% of the power pad ?
    We understood that there is no problem if we mounted the OPA561 with the normal reflow profile.
    However, if we should mount the OPA561 with 95% solder coverage, they may be change the mounting method...

    Best Regards,
    Ryuji Asaka
  • Asaka San,

    I am pleased to hear that connecting the OPA561 PowerPAD to V- corrected the oscillation.

    The OPA561 can operate at a high current level relative to the small package size. Heat is conducted away from the OPA561 die through the underside PowerPAD and the many package leads. It is imperative that PowerPAD connection meet the PowerPad assembly and PCB requirements outlined on OPA561 data sheet pages 10 and 11. The expectation is there will be a thorough solder connection between the OPA561 PowerPAD and the copper plane that dissipates the heat. That means the solder coverage should come as close to 100 % as possible.

    If the OPA561 is operated at a significantly lower power level than its maximum, them it may be possible to use a 65 % solder coverage. It is a matter of the die temperature attained in the application when it is operating. If the die becomes too hot it will go into thermal shutdown to protect itself.

    Regards, Thomas

    PA - Linear Applications Engineering