• State Resolved
  • Locked Locked
  • Replies 15 replies
  • Answers 2 answers
  • Subscribers 48 subscribers
  • Views 1859 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

OPA548: Voltage extension circuit

chentsu chung
Mastermind 6180 points

Part Number: OPA548
Other Parts Discussed in Thread: THS3091, THS3491, OPA454

Hi Team,

I'd like to use the voltage extension circuit by OPA548 for +/-50V solution, the spec is:

+VS: +45V

-VS: -45V

Vout: -40V to +40V

Gain: about 4X to 5X

For gain adjustment, I can modify R2 and R3, but is there anything I need to modify for the spec?

Thank you,

C.T.

  • 0
    Guru 48395 points

    Hello CT

    You say this is a OPA548 but show THS3091 high speed current feedback amplifiers in the schematic.?

    What speed do you need, If it is in fact high speed, use the new THS3491

  • 0 in reply to Michael Steffes
    Mastermind 6180 points

    Hi Michael,

    The circuit I posted is provided from TI, but the voltage range of THS3091 and THS3491 is +/-15V.

    As I mentioned, the spec is: 

    +VS: +45V

    -VS: -45V

    Vout: -40V to +40V

    Gain: about 4X to 5X

    THS3091 and THS3491 can only provide +/-30V range in this ciscuit, so that is why I ask for OPA548.

    Please help to check if use OPA548 in this circuit that TI provided for our spec, what should I modify?

    Thank you.

    C.T.

  • 0 in reply to chentsu chung
    Guru 48395 points

    Again, answers can only be made in the context of your desired speed or bandwidth and slew rate. Also what is the load? 

    I personally have always been dubious of that circuit as it seems like a direct path to oscillation. Recall that normally devices recommend supply decoupling caps - where are they here? If you add them, then the op amps bootstrapping the supplies must drive those at the speed of your output signal, 

    Where did you find that circuit, was there measured waveforms? 

  • 0
    TI__Guru* 93105 points

    Hello Chentsu,

    The circuit you propose is very similar to one shown in Figure 67 of the OPA454 HV Op amp datasheet. You can reference the OPA454 datasheet for more information about the circuit.

    http://www.ti.com/lit/ds/symlink/opa454.pdf

    Since the peak output current in your application is 45 V/ 100 Ohms, or 450 mA, you may be able to use a single OPA454 HV Op amp with some added external output transistors to accomplish the high-voltage, high-current task. The OPA454 can be powered directly from +/-45 V supplies, and then the external transistors would drive the load. Figure 66 in the OPA454 datasheet shows the configuration called "External Output Transistors Boost Output Current Greater than 1 A."

    You may find that the OPA454 circuit configuration using the external transistors may be an easier to apply than one using three OPA548 power Op amps. And the overall cost should be lower as well. Do be sure that you whatever circuit you use meets the application bandwidth and slew requirements as Michael has pointed out.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    Mastermind 6180 points

    Hi Michael, Thomas,

    Please check the application note that provided from TI: HighOutputVoltageExtensionConcept.pdf

    And the require slew rate is >80V/us @no load

    Output current: no sure, needs to be tested, but I'm sure that OPA454's 50mA is too small

    Maybe the solution I plan to do is not good, or can I ask, for such spec, what's the best solution?

    Application: LCD tester

    +VS: +45V

    -VS: -45V

    Vout: -40V to +40V

    Gain: about 4X to 5X

    slew rate is >80V/us @no load

    Output current: no sure, needs to be tested, but I'm sure that OPA454's 50mA is too small

  • 0 in reply to chentsu chung
    TI__Guru* 93105 points

    Hello Chentsu,

    Unfortunately, the slew rate of our power Op amps is much lower than 80 V/us. Therefore, I do not have a device I can recommend for the application circuit.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    Guru 48395 points

    So you say an LCD load, that is capacitive - the required peak output current is the I/C = dV/dT. what is the C?

  • 0 in reply to chentsu chung
    TI__Guru* 93105 points

    Hello Chentsu,

    Since we do not offer a Precision Amplifiers Op amp that can meet the high-voltage, high-current, and high slew rate required by your application I suggest checking the Apex Microtechnology web page; specifically, their Power Op amp Selector tables. It appears they have products that will cover all three of your requirements.

    TI has no affiliation with Apex Microtechnology whatsoever and we have no responsibility for any outcome should you decide to employ their products. I mention them to assist you in finding a solution for the applications inquiry you made to TI.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    Guru 48395 points

    Yea, Tom is correct and I was heading that way if you could provide a Cload which you will need to arrive at to eventually select a solution. 

  • 0 in reply to Michael Steffes
    Mastermind 6180 points

    Hi Michael, Thomas,

    Thanks for your great support.

    Actually, the project used APEX solution, and TI provide the application note below to replace it

    3323.HighOutputVoltageExtensionConcept.pdf

    But looks like this way from the application note can not be used for the updated spec.

    Thank you very much!

    Best regards,

    C.T.

  • 0 in reply to chentsu chung
    TI__Guru* 93105 points

    Hello Chentsu,

    Yes, it does appear that the +/-40 V output voltage you require cannot be provided by the THS3091/THS3491 circuit shown in the paper. The best solution does seem to be one that uses an Op amp that provides the high-voltage, high-current and high slew rate required by the application.

    Sometimes TI doesn't have a device that meets all the applications requirements. Nevertheless, thank you for giving us the opportunity to consider your application and how we might best satisfy it with our products.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    Guru 48395 points

    yes Chentsu, 

    And I am still waiting for a cload to take this further, which I can.

  • 0 in reply to chentsu chung
    Guru 48395 points

    Again C.T.,

    If you can provide the load capacitance, I might be able to take this further. 

  • 0 in reply to Michael Steffes
    Mastermind 6180 points

    Hi Michael,

    Thanks for your great support, now I can only get these spec information. Our customer can't make sure the spec now. 

    Thank you,

    C.T.

  • 0 in reply to chentsu chung
    Guru 48395 points

    Hello CT. 

    few questions here still, 

    1. You say no load, but it is likely a capacitive load - that becomes a problem at high swings and for stability - do you have a typical value

    2. You say the current solution is an Apex part, what part and why do you want to replace it (for other projects, I have been importing Apex models into TINA quite a lot lately - some are ok, others not so much. But if I had a part number and circuit, I could try to construct in TINA

    3. Do you know who did that TI app note on bootstrapping the supplies, having that, I would be asking if there was any bench data. Using the THS3091 sounds like an older app note - the newest in that type is the THS3491. Much higher performance, but not the swing you are asking for.