• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 2 answers
  • Subscribers 47 subscribers
  • Views 1574 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.

Original question:

OPA182: Need some help with an OP AMP in my circuit!

OPA182: single ended to differential output with OPA182?

Chris van der Aar
Intellectual 570 points
Part Number: OPA182
Other Parts Discussed in Thread: OPA2182, REF102, OPA2192, ADS8588S, ADS8688

Tool/software:

For a new project we are starting I have to design an interface PCB for a flux sensor (Hall sensor) that has to interface with an external analog I/O module with -20V..+20V differential input range.

For the flux sensor I need an accurate current source of 2mA. Therefore I want to use the REF35300 with opamp feedback like described in SBOA046 (page 6 at https://www.ti.com/lit/an/sboa046/sboa046.pdf)

I need 2mA current so I can use a 3V ref and 1k5 low TCR resistor. Furthermore I searched for an opamp with  low offset (the OPA111 is very old in this example) and I think the OPA182  (or OPA2182 because I also need  an extra opamp for inverting my ref voltage for offset compensation of my instrumentaiton opamp) could be a could candidate for this.


Furthermore the amplified output of the instrumentation amplifier (-10V..+10V) that amplifies the diff voltage of the flux sensor that acts like a kind of wheatstone bridge needs to be converted from single ended to differential (I will use a low noise power supply of +12V/-12V for this).

Do you think I can also use the same opamp (OPA2182 so i have dual version) for this task where I use 1 opamp as unity gain buffer for the positive side and the other as unity gain inverter for the negative siude? What is the max capacitive load for this opamp with unity gain amplifying?  Or do you suggest another type of opamp for this task? Our bandwidth will be max 20kHz so I think that is not a problem for most opamps. But the opamp must operate at +12V/-12V dual supply because of the analog range. The output of my flux sensor interface PCB has to be connected to the input of the external analog I/O module via  cable (I guess 0.5mtr max) with the following analog input specs:

  • 0
    TI__Mastermind 24788 points

    Chris,

    1. I think you are designing a few different applications and you want to use the OPA2182 on both these applications.  One application uses the current reference illustrated in the OPA111 data sheet.  The other application is a single-ended to differential conversion.  Please correct me if I am missing something.
    2. The OPA2182 is a chopper op amp.  This type of op amp has very low input offset voltage (Vos) and very low Vos temperature drift.  In the case of the current reference, the offset voltage will introduce an error as it will add to the reference voltage and will slightly change the current Iref = (Vref + Vos)/R1.  Thew image shows REF102 (10V reference). As you mentioned this is an old image, so you may want to consider a new reference option.  In any case if the reference voltage is 10V, a 5kΩ resistor would be used to set the current.  For this reference an op amp Vos of 1mV would correspond to a 200nA current error which is 0.01% of 2mA.  The OPA2182 has a max Vos of 4μV which corresponds to  0.00004% error (0.8nA).  The Vos drift will also be quite good on the OPA2182.  The key point here is that other error sources will likely dominate and you could probably assume the OPA2182 error is basically zero.
    3. In general, I do think that a chopper amplifier is a good choice for a precision DC current source.  My only concern would be that the chopping technique does introduce a current transient on the inputs due to charge-injection and clock-feedthrough.  Optimizing Chopper Amplifier Accuracy covers the advantages and limitations of chopper amplifiers.  I do not think the chopping transients would be very significant compared to the 2mA DC current, but I just want you to be aware of them.  As an alternative, you could consider a trimmed CMOS op amp like OPA2192 as an option.  This device has a maximum Vos of 25uA, so it is not quite low as teh OPA2182 but it still achieves an error of 0.00025%.
    4. Single-Ended to Differential Using a Two Op-Amp Circuit provides an example for using an op amp as a single ended to differential conversion.  Very often this topology is used to drive an ADC.  If you are using this to drive an ADC you should make sure that the op amp you choose has the dynamic response needed for your ADC.  Some ADC have transients on their input that the op amp needs to respond to.  For example, on a SAR ADC the sample and hold circuit creates a transient current demand for each conversion to charge the ADC internal RC circuit.  I don't know what your single-ended-to-differential circuit is connected to so it is hard to say if the OPA2182 is a good choice.  Theoretically, there is no issue with using the OPA2182 for this case.

    I hope this background helps.  I think the OPA2182 is a good choice for precision DC applications, and that may fit your needs well.  Alternatively, a non-chopper device that doesn't have the chopping transients, but is not quite as low Vos and Vos drift may be a consideration (OPA2192).

    Best regards, Art

  • 0 in reply to Art Kay
    Intellectual 570 points

    Hi Art,


    Thanks for your clear information and I think I will choose the OPA2192 just to be sure that we do not get issues by current transients and the accuracy is still good enough for our application.
    Concerning the diff output amp I assume that I can also use the OPA2192 for this? I found a similar question here but in my case I already have -10V..+10V range out of my instrumentation amp so I only need 2 unity gain buffers at the output where one is acting as inverting amp. Or I can choose to operate my instrumentation amp at single supply so I get a 0..+10V single ended at the output of the amp and making a differerential -10V..+10V at the output like this example (so I need a 10V ref)  but I do not think it has advantages? My instrumentation amp can operate both at single as dual supply. Maybe you have a suggestion for this? 

    Single ended to differential conversion for ads1278 - Amplifiers forum - Amplifiers - TI E2E support forums

    The ADC is integrated in an external I/O module that we buy (as part of a motion control system) and according the datasheet of that module that are using the AD7609 which is a SAR ADC but it has LPF filtes and transient clamps at the input so maybe the transient current at the input is minimal with this external module? See below the AD7609 func block diagram and the input specs of the module. We need a cable to connect our PCB with this module but I think it will be not extreme long (<1mtr I expect). The max capacitive laod of the OPA2192 is 1nF according the datasheet so I think that should not be a problem. Besides our bandwidth is limited to 15 kHz (I use a Sallen Key filter where I want to use this same opamp as second order filter for a filter at 15kHz. The advantage is that  I can use one opamp type for the whole design. And maybe we can use this one for other projects when we have dual supply and need low offset and low drift?

  • 0 in reply to Chris van der Aar
    TI__Mastermind 24788 points

    Chris,

    Yes the OPA2192 will work for the single ended to differential conversion. You could also use a fully-differential-amplifier (FDA) as was shown in the E2E post you mention.  Single-Ended to Differential Signal Conversion Using an Op Amp and FDA for Unipolar Signals is a document that covers the conversion using a FDA.  Often it is important to carefully select the op amp or FDA to make sure it has the transient performance to drive the ADC input.  This is because many SAR ADC have a sample-an-hold RC circuit on the input that generates a transient that the amplifier needs to settle for.  In the case of AD7609 the ADC has a 1MΩ resistive input impedance and an amplifier.  The sample and hold is internal and driven by the internal amplifier.  Thus, the  amplifier that drives the ADC doesn't need to respond to any transients, and you only need to consider your signal conditioning requirements when selecting the amplifier.  In other words, you choose the amplifier bandwidth, slew rate, noise and other parameters according to your signal requirements and don't need to worry about the ADC.  Note that the TI equivalent ADC to AD7609 is ADS8588S.  There is a series of ADC that have a similar input structure: ADS8588, ADS8688 and variants.  In general, there are many options that can work for doing the single ended to differential conversion.  The two documents I gave links to should help you with component selection.

    I hope this helps.

    Best regards, Art