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.

OPA2145: amplifier and other component selection

Part Number: OPA2145
Other Parts Discussed in Thread: LF412, , TL082H, LOG114, TL072, OPA397, OPA392, OPA197, OPA192, LMP7721, OPA828, OPA2392, LMP2012

Hi Experts,

We are working on a solution with log amplifier and thinking of developing a circuit  based on the design from application note here -

Amplifier used/recommended in this design is LF412. Since it is not available we are looking for alternatives and these amp have similar properties - OPA2145, TL082H. Can we use anyone from it or any other part will suit much better ? Our sense current range is 100pA to 10mA. Supply voltage is +5V. Required Output 0-2.5V.

LOG114 meets our exact requirement but it is not available all the time in the market so we want build the circuit with commonly available opamp.

This circuit also requires a special 1K resistor "*1% film resistor †1 kΩ (±1%) at 25°C, +3500 ppm/°C." where we can get it ?

  • Hi Tylor,

    I have built this circuit with the TL072 many years ago. But any good JFET- or CMOS-OPAmp will do.

    Is your circuit powered by a bipolar supply voltage? Is the input offset voltage critical? What is the bandwidth?

    This circuit also requires a special 1K resistor "*1% film resistor †1 kΩ (±1%) at 25°C, +3500 ppm/°C." where we can get it ?

    You may find this thread useful:


  • Hi Kai,

    Thanks for you response !

    I am not familiar with analog circuits. May i know what the term refers "good" ? what parameters decides an amplifier as good?

    we will use this photodiode -

    Yes, we have bipolar power supply of +15 to -15V.

    I did check the reference for the TCR but adapting the same with our requirement is confusing. Can you please give additional details/information? 

  • Hi Tylor,

    May i know what the term refers "good" ?

    By "good" I mean an OPAmp with very little input offset voltage. I'm sure that a TI's expert from this forum will recommend you suited OPAmp.

    I did check the reference for the TCR but adapting the same with our requirement is confusing. Can you please give additional details/information?

    You have to adapt the 15k7 / 1k resistor divider when using a PT1000 element instead of the 1k / Q81 resistor. I would replace the 1k / Q81 resistor by a PT1000 element and the 15k7 resistor by a series circuit of 17k2 and 28R (E192 series, for instance). This gives a good match at 25°C (17228R / 1097.35R = 15.6996). The temperature coefficient of PT1000 at 25° is about +3542ppm/°C, when you look up the resistance values of PT1000 element from an official table. This should be close enough to the +3500ppm/°C of 1k / Q81 resistor.


  • Hi Tyler,

    Thank you for your inputs, Kai. As far as the op amp selection goes, depending on the temperature range, you may need to consider maximum input bias current, IB.  Since most of the IB in CMOS and JFET op amps is a function of the current leakage of reverse-biased ESD protection diodes, it double every 10 deg C;  thus, typical IB of 1pA @25C becomes1nA @125C (factor of 2^10).  Since you state the need for sensing current down to 100pA, IB vs temp change must be taken into account if you operate above the room temperature.  Also, the input voltage offset (Vos) and rail-to-rail input/output voltage range adds flexibility to any design.  All and all, I believe the best choice for a single 5V supply would be OPA392 or OPA397 while for +-15V dual supply OPA192 or OPA197 would be my choice - see below.

    For even lower IB and noise, you could also consider LMP7721 (5V) and OPA828  (+/-15V) but they require much higher quiescent current, IQ.

  • Hi Kai,

    Thank you for the detailed explanation !

    We are planning to use this PT element - Do you see any issues in using this part?. plz recommend if you find any better alternative.  It has 3850ppm/°C. May i know how you've calculated the temp coefficient for 25°C and  got this value "+3542ppm/°C". With this formula ?

  • Hi Marek Lis,

    Thanks for the info and recommendation on opamp. Now we have better understanding with parameter selection of opamp. Apart from "Input bias" and "Offset Voltage", do we need to consider anything else? Unfortunately OPA392 and OPA397 are not available. so we need to choose one from the available product.

    We want the opamps to be powered with +5V single supply. It is better if it is dual opamp in single chip to avoid space consumption.

    Based on the recommended parameters we shortlisted some, Do you recommend any part from this or other parts?

  • Tylor,

    I see 30,000 available inventory of the dual channel OPA2392.  Additionally, you may consider three other parts - see below:


  • Hi Tylor,

    for a PT1000 element the temperature coefficient is not constant but varies slightly with temperature. Because of that I center the calculation arround the mean temperature of your application which I assume to be room temperature (+25°C) and consider only a small temperature difference:

    alpha = (R26 - R25) / R25 / 1°C

    From an official table one can read out the following resistances of PT1000 element:

    +25°C: 1097.3 Ohm

    +26°C: 1101.2 Ohm

    which yields:

    alpha = (1101.2 - 1097.3) / 1097.3 / 1°C = 0.003554 = +3554ppm / °C.

    The +3542ppm/°C number given earlier by me was calculated with the help of a fast approximation formula. As you can see the difference between both results is very small, especially when you keep in mind that the temperature coefficient of such a temperature sensistive resistor can show manufacturing tolerances of several 100ppm/°C.


  • Hi Lis,

    For me the portal shows OPA2392 is out of stock and also it is BGA type package which increase complexity in proto testing so decided to go with LMP2012. Thank you for your time, kind response and recommendations !

  • Hello Kai,

    Thanks for the detailed explanation ! All the provided information are very helpful in designing this circuit. Now it is complete. Would you be able to review it ? as you have already built same circuit with TL072 chip, your review will add more values. If so Can I share it in mail or in personal chat ?

  • Hi Tylor,

    as I'm not a TI's employee you may want to send the schematic to Marek.