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.

OPA657

Other Parts Discussed in Thread: OPA657, OPA659, OPA847, THS4211, THS4271

What is the difference between the packing of OPA657U and OPA657UB? I want to order sample IC first, which one should I order?

  • The OPA657UB is a High Grade version of the OPA657. It has better DC specs compared to the OPA657U. Please see page 3 & 4 of the datasheet to compare the spec variation between the standard and high grade. If you care about dc offset and drift over temperature then the High Grade offers superior performance.
    -Samir
  • Thanks for the reply.. I want to design a low noise high speed amplifier circuit. For this which ic will be better opa657 or opa659?

    1) opa657 has got lesser noise and higher bandwidth but i am not very much clear about the slew rate difference. opa657 has got 700 V/uSec for gain of +10 V/V, while opa659 has got 2550 V/uSec for gain of +1 V/V. Which one out of these will be better for gain of +10 V/V or even more?

    2) In case of opa659, bandwidth is defined as 35 MHz for V0 = 200mVpp, G = +10 V/V, does it mean that for higher output voltage (V0), bandwidth will be even lower?

  • Hello Nishant,
    The OPA657 will have superior noise and Gain-BW performance compared to the OPA659. Note, however the OPA657 is not unity gain stable and needs a minimum gain of 7V/V to be stable. The OPA847/OPA847 opamps have lower noise and BW compared to the 657/657. They however have a bipolar input stage, rather than JFET. To answer your other questions:
    1. The Slew Rate relationship to Gain is similar to the Gain-BW relatioinship. Higher the gain, lower the slew rate.
    2. For small signal the Gain will stay fairly constant with output swing. Howevever as the signal gets larger (4Vpp and above), you will start to see a degradation in BW due to slew rate limitations. These effects are well modeled in the Spice models which are available on the product webpage under the "Tools & Software" tab.
    -Samir
  • Hi Samir,

    opa847 has good noise level and BW but it has quite higher input bias current. Since my application is to detect low current, so i can not go with that.

    As you mentioned, that for good stability opa657 should be used for gain > 7V/V. Since we want to use it for higher gains (10V/V) then there should not be any problem regarding stability too. So can we say that opa657 will be better than opa659 for this application? My only worry is about slew rate. Still i don't get how to compare slew rate of these two, which is better for gain >= 10V/V.

    If we want to get gain >1000, can we go for two stages of opa657 with gain 40 V/V each?

  • The OPA657 will be more suitable for the application. As I mentioned previously, when you increase the amount of gain, the slew rate is going to go down, similar to a Gain-Bandwidth type of relation. Again, please use TINA spice to simulate this.
    Yes, you can have Gain =40V/V, BW will scale down accordingly.
  • At high frequency, d.c. power supply also adds noise in the output voltage signal. Can you suggest some IC which can give high stable d.c. power supply (6 V) required for these op-amps?

  • Hello, The Power side is not my specialization. Please post your questions on the Power Mgmt. forum (link below) for best support.
    http://e2e.ti.com/support/power_management/linear_regulators/
  • Dear Samir,

    I have used OPA659IDRBT product which has SON(DRB) type packaging and using that I got fabricated a PCB board.  Now, I want to use OPA657 on the same PCB replacing OPA659. For that I ordered OPA657UB and OPA657U but they have different packaging type i.e. SOIC(D). Kindly suggest me, which product code I can use for opa657 on the same board.

    -- Nishant

  • Hi Nishant, The OPA657 is not available in that package style. Can you please let me know why you are considering OPA657 over OPA659 and what your design requirements are and I can look to see if there is an alternate part in the same package.
    -Samir
  • Dear Samir,

    I have checked that for higher gain (>10) opa657 has fast responce time (rise and fall time) as compare to opa659. I want to measure a low voltage signal (< 50 micro volt). So i want to design an amplifier for that which can respond very fast (< 1 ns). Kindly suggest me.

  • Hello Nishant, Unfortunately with the package style you have chosen, we don't have an alternate amplifier to be a drop-in replacement with higher bandwidth that also has JFET inputs. You can try using the THS4211 and THS4271 both of which have higher bandwidth that the OPA659 but have bipolar transistor inputs and so will have a higher bias current. The noise will also be lower for the THS devices, especially the THS4271.
    -Samir
  • Dear Samir,

    I have seen in opa659, that although i apply 6V at Vcc, opamp saturates at 4 V. Is there any specific reason behind this?

    For opa657, i can see that there are two operating ranges in terms of output voltage i. e 200mVpp and 2Vpp. So what is the safe range of output voltage so as to avoid saturation of the opamp?

  • The OPA659 has an output swing limitation of 2V from the positive rail when driving a 100Ohm load. Please see Page 4 of the datasheet for this spec. Amplifiers usually have some kind of limitation like this. Even amplifiers classified as rail-to-rail may only swing to within 200mV of the rail in some cases. As the output load increases it is more difficult to get close to the rails.

    The OPA657 specifications are on page 3 of the datasheet. It is 1.5V from the rail on the output side under 100Ohm load. There will also be an input swing limitation (under "Input" characteristics of the electrical table) that may be important depending on the amplifier configuration.

    Please also look at the following training to better understand amplifier specifications and how they work. You may find this quite useful as a refresher.

    -Samir

  • In opa659, decoupling capacitors connected with Vcc are parallel combination of 0.1uF and 10uF while for opa657, it is 0.1uF and 6.8uF. Why? How to decide these values?

  • Hi Nishant,
    The exact value of capacitors do not matter too much. What we are interested in is having a low impedance power-supply path across frequency. In order to achieve this we use several caps in parallel since a single cap cannot achieve this. The large value caps tend to be inductive at higher frequencies so we parallel them with smaller value caps. A general rule of thumb is to have a capacitor in the range of 1-10uF common across multiple devices. Then use a 10-100nF cap right next to the DUT pins. I usually add an extra 1nF-100pF cap as well if space allows.

    -Samir
  • I am designing a high frequency amplifier, do these capacitors affect the speed of amplifier? For high frequency circuits which type of capacitors should we use like ceramic, tantalum or something else if you can suggest..
  • The capacitors will not affect the speed of the amplifier but will affect its stability and linearity performance. I stick with ceramics, X5R and X7R. For < 1nF caps I try and use C0G/NPO caps where possible.