• State TI Thinks Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 48 subscribers
  • Views 463 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.

OPA593: Supply

Ray Vincent Aranzaso
Mastermind 18180 points
Part Number: OPA593

Hi,

Good Day. I have a customer who is working with OPA593. Please see below his query for your reference. Thank you very much.

This is a rail to rail op amp, XOPA593DNTR, and I am required to go from 0 to 5V. Meaning I have to meet those rails with no suppression.

Usually rail to rail op amps have some suppression, meaning they do not quite get to the rails.

I am in the middle of choosing a power supply to put in the schematic that will work with this op amp.

To get to 5V I plan on using a +15V power supply. If I use a +5V power, I do not think the output will get to +5V due to suppression, correct me if I am wrong.

I have to get to exactly 5V and 0V.

To get to 0V would I need a negative power supply, or would just a positive power supply referenced to ground (0V) get me to exactly 0V?

With my strict requirements I need to be able to swing from 0V to 5V and I am trying to figure out what rails I need connected to the op amp XOPA593DNTR.

I just have to put power supplies on this new schematic and understanding what the op amp needs for rails to swing from 0 to 5V with no suppression is critical. I have to reach 0V and 5V.

Best Regards,

Ray Vincent

  • 0
    Guru 140200 points

    Hi Ray,

    are you sure the OPA593 is meant? This is no rail to rail OPAmp. The input voltage would need stay 3.5V away from the positive supply voltage. And the output voltage swing heavily depends on the load current. Also, the minimum supply voltage would be 10V.

    For the worst case scenario (voltage buffer and a load of 600R) a supply voltage of +8.5V and -2.5V would be sufficient. More can only be said when seeing the schematic.

    Can you show a schematic? And some more details on the application would be helpful (load, bandwidth, source impedance, etc.)

    Kai

  • 0
    TI__Guru* 93105 points

    Hi Ray,

    As Kai mentions the OPA593 is not classified as a Rail-to-Rail op amp and both the input common-mode voltage (Vcm) range and output voltage swing range (Vo) are limited in terms of useable range relative to the supply voltage settings. Additionally, the Vo specification - even though it may indicated the output swings to within tens or hundreds of millivolts of the supply rails - is not an indication of the linear Vo range. It is a hard stop at the rail swing limit and the op amp output is not operating linearly at that point. You mention that you have a strict requirement for a 0 to 5 V swing and that implies high linearity must be maintained.

    You should be able to easily achieve the +5 V input and output using a +15 V supply because the OPA593 performance is within the Vcm and Vo specification ranges. It is the 0 V used on V- that will be the problem. As you have suspected V- needs to be somewhat below 0 V to maintain the output within the linear operating range. The most direct approach is to include a V- supply. It doesn't have to be -15 V, but a few volts below 0 V.

    The OPA593 open-loop gain (Aol) specification provides information about the linear output voltage range for specific load conditions. I don't know what load current you need from the OPA593, but the the Aol specification shows the linear Vo swing range for different load resistances. I would plan on using at least -3.3 V for V-, or -5 V to be sure.

    Certainly more details about your OPA593 application would be helpful.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    TI__Mastermind 18180 points

    Hi Thomas and Kai,

    Good Day. Please see below the response of our customer to your replies. Thank you very much.

    The OPA593 is a rail to rail op amp. Please check your datasheet, which I attached. OK I have answered one of my own questions from the datasheet. With an output current of 50mA the voltage output swing from the rail is maximum 125mV. So I do need a power supply greater than +5V to ensure my output voltage reaches +5V. OK so now the question is with this being a rail to rail op amp, do I need a negative supply to ensure the output voltage reaches 0V.

    opa593.pdf

    Best Regards,

    Ray Vincent

  • 0 in reply to Ray Vincent Aranzaso
    TI__Guru* 93105 points

    Hi Ray ,

    TI tends to reserve the use of describing an op amp as a "Rail-to-Rail op amp" to when both the input and output ranges are rail-to-rail. In the case of the OPA593 the datasheet states that the just output stage is rail-to-rail. I suppose it comes down to semantics.

    Despite the OPA593 being described as rail-to-rail output Vo can only so close to the supply rails, like all op amps. And as the OPA593 Aol specifications indicate the more heavily loaded the output is from a current output requirement, the less close it can swing to the rails. Since the OPA593 datasheet is "Preliminary" we don't have full information to work with yet.

    You have the requirement to maintain high linearity as I mentioned previously. A +5 V supply won't provide any margin when the output must achieve 5 Vpk. It is probable that the supplies would need to be at least 2.5 V beyond the output expected swing range of 0 to 5 V. I suggest supplies of +7.5 V and -2.5 V at the lowest. I confirmed this with the designer and he said that should be about right.

    Regards, Thomas

    Precision Amplifiers Applications Engineering

  • 0 in reply to Thomas Kuehl
    TI__Mastermind 18180 points

    Hi Thomas,

    Good Day. Another query from the customer, please see below for your reference. Thank you very much.

    If you notice on the first page of the datasheet under features it says Rail to Rail Output. After you dig into the datasheet you find that not to be true. I realize this is a very new part. You may consider removing the reference to Rail to Rail Output on the first page of the datasheet under features.

    I just have a couple more questions. Where it says voltage output swing from the rail, depending on output current, does this pertain to both the V+ and V- rails or just the V+ rail?

    In the linear operation range, the minimum common mode voltage is (V-) - 0.1. So if my V- is 0V it implies it can handle a -0.1 volt input and still be in the linear range. But since I would not have a negative rail, I don't think it can handle a -0.1 volt input. Do you agree? In the end I have come to the same conclusion you did in that I need a negative rail.

    After thinking about it because this is a CMOS output where it says voltage output swing from the rail, it is more than likely pertaining to both the V+ and V- rail because of the CMOS output (2 FETS)

    Best Regards,

    Ray Vincent

  • 0 in reply to Ray Vincent Aranzaso
    TI__Guru* 93105 points

    Hi Ray,

    Where it says voltage output swing from the rail, depending on output current, does this pertain to both the V+ and V- rails or just the V+ rail?

    The op amp Vo specifications are commonly specified for dual supply operation. Where the upper output CMOS transistor sources current to a load connected to ground, and the lower CMOS transistor sinks current from ground depending on the output polarity. Therefore, the swing is applicable to each output. 

    In the linear operation range, the minimum common mode voltage is (V-) - 0.1. So if my V- is 0V it implies it can handle a -0.1 volt input and still be in the linear range. But since I would not have a negative rail, I don't think it can handle a -0.1 volt input. Do you agree? In the end I have come to the same conclusion you did in that I need a negative rail.

    Yes, your understanding is correct.

    After thinking about it because this is a CMOS output where it says voltage output swing from the rail, it is more than likely pertaining to both the V+ and V- rail because of the CMOS output (2 FETS)

    Yes!

    Regards, Thomas

    Precision Amplifiers Applications Engineering