Because of the holidays, TI E2E™ design support forum responses will be delayed from Dec. 25 through Jan. 2. Thank you for your patience.

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.

OPA462: How to Calculate the Power Consumption of Operational Amplifiers in the Application of Howland CurrentPump

Part Number: OPA462
Other Parts Discussed in Thread: OPA365, , OPA593, OPA192, OPA455

HI bro

  in my case,the OPA is very hot,the System can run only few minites and the the OPA will broken。so can you tell me How to Calculate the the  Power Consumption of Operational Amplifiers? I'ii check my sch,thanks so much.

  • sorry ,R8 is 1k in fact ,and DAC set the load current is 10 mA

  • Hi Wenhe, 

      in my case,the OPA is very hot,the System can run only few minites and the the OPA will broken

    The simulation is shown OPA365 and topic is shown OPA462. The OPA462 is high voltage power amplifier, while OPA365 is low voltage op amp (up to 5.5Vdc, not +/-35Vdc as shown in the simulation). So I am not certain which op amp you are interested. 

    Enclosed is an example of OPA593, which is +/-42.5Vdc and +/-250mA power amplifier, which it should meet the above constant current requirements (maybe the BW as well). 

    Iout is derived from Vset/Iout, which RL is configured for 50Ω in this case. The difference amplifier has gain of 1V/V and the feedback resistor ratios should be well matched better than +/-1% (prefer 0.1% or better for the higher CMRR performance). 

    I used OPA192 as the feedback buffer. You may not need to, if the feedback errors are small, but you have to use large value feedback resistors (it may increase the system noises of the circuit). You may use lower cost, lower Vos op amp and/or zero drift op amp for the buffer, which we have many of these to choose from.  

    OPA593 Howland PWM +-100 mA Ipump PD 09202023.TSC

    In terms of OPA365 is overheating, perhaps the supply voltage rails is too high. OPA365 is only rated up to 5.5Vdc, but it should be able to source up to +/-10mA without issues.

    The power dissipation calculation is shown in the OPA593's simulation. 

    Vcc_diff * Icc + Vee_diff*Iee + Iq*(Vcc-Vee)

    OPA593 still has PSpice modeling issues when it is dealing with the simulation in source/sink current of the Vcc and Vee, so I am unable to demonstrate the calculation via simulation. 

    Please let me know which op amp you are interested in and I will simulate it for you. 

    Best,

    Raymond

  • Hi Raymond,

     thanks for your reply, I use OPA462 in fact,I have some confusion about your formula,

    1. How could I get the Icc and Iee without simulation, Its depond on the opa or not?
    2. Like the power consumption calculation method of LDO is equals the input power minus the output power. Here, I can approximate that the opa power consumption=Vcc_diff * Icc + Vee_diff*Iee + Iq*(Vcc-Vee)-Iload*Vout,am right?if not,pls tell me the where the problem is.

    Thanks so much.

    BR

    Aape

  • BTW, Can you recommend a suitable circuit and OPA ,It can generate a variable stable current of 1-10mA within  1k-10k load. I hope the response time is within 10uS.

  • Hi Wenhe,

    Can you recommend a suitable circuit and OPA ,It can generate a variable stable current of 1-10mA within  1k-10k load.

    Yes, you would need to have high voltage power amplifier if you are using Howland Current Pump to generate the 1-10mA constant current. 

    With 10mA*10kΩ , 100V will drop across the load, and the supply voltage has to be higher than 100V + operating overhead. OPA455 or OPA462 should work to generate the constant current per the Howland Current Pump requirements.   

    https://www.ti.com/amplifier-circuit/op-amps/power/products.html?login-check=true#1261max=45%3B180&sort=1261max;desc&

    Let me generate the constant current file and we can proceed from there. 

    Best,

    Raymond

  • Hi Wenhe,

    I use OPA462 in fact

    Here is the simulation example for OPA462. OPA455 should work as well. 

    OPA462 has the slew rate at approx. 32V/usec. For 100V step, it will take approx. 100V/32V/usc = 3.125usec minimum. Please aware of the rise/fall time limitation. 

    OPA462 Howland PWM 1-10 mA Ipump 09212023.TSC

    How could I get the Icc and Iee without simulation, Its depond on the opa or not?

    The OPA462's quiescent is approx. 3.2mA at +/-90V supply rails. The additional Icc in DC can be estimated by Vset/500Ω = 5mA, if Vset = 2.5Vdc. So the total Icc is approx 5mA + 2*Iq, in this case Icc ≅ 11.4mA --> simulation indicates approx. 11.4mA (2*Iq due to 2X of OPA462).

    Icc depends on the load and Iq in the configuration. 

    Here, I can approximate that the opa power consumption=Vcc_diff * Icc + Vee_diff*Iee + Iq*(Vcc-Vee)-Iload*Vout,am right?

    When I wrote out the above equation, these variables are function of time in simulation. The total power dissipation in OPA462 depends on the output waveform. It should be an average power for AC signal, where the average power is integrated over time (not peak power or rms power). 

    Note: Icc or Iee current during simulation include Iq as well, because the Iq current is simulated in our PSpice model. Depending on the configuration, you may have to add or subtract from the total current, if you want to be accurate.

    Overall, the AC power via heat dissipation can be calculated via the following equation.  

    BTW, there is no negative dissipated power. The sign indicates that it is sinking power, but it is still the heat built up inside of the IC. 

    If you have other questions, please let me know. 

    Best,

    Raymond

  • Hi,RAY

        It's really great to discuss OPA issues with you,I have some confusion as below,

    Raymond Zhang1 said:
    When I wrote out the above equation, these variables are function of time in simulation. The total power dissipation in OPA462 depends on the output waveform. It should be an average power for AC signal, where the average power is integrated over time (not peak power or rms power)

    When I wrote out the above equation, these variables are function of time in simulation. The total power dissipation in OPA462 depends on the output waveform. It should be an average power for AC signal, where the average power is integrated over time (not peak power or rms power)

    1. I want to get the OPA Power consumption under the ioad current is constant 10mA(the current donest change),in this case ,the equation Popa=Vcc_diff * Icc + Vee_diff*Iee + Iq*(Vcc-Vee)-Iload*Vout  is right?and the Icc or Iee can get only with simulate or real test? Can I get it through other ways。
    2. another question,in your sch,R6 +R4≠R5,its ok ?

    Thanks a lot 

    BR

    Aape

    When I wrote out the above equation, these variables are function of time in simulation. The total power dissipation in OPA462 depends on the output waveform. It should be an average power for AC signal, where the average power is integrated over time (not peak power or rms power)
  • Hi Wenhe, 

    Enclosed are some of Howland current pump references. 

    https://www.ti.com/lit/an/sboa436/sboa436.pdf?ts=1695365929519&ref_url=https%253A%252F%252Fwww.google.com%252F

    https://www.ti.com/lit/an/sboa437a/sboa437a.pdf?ts=1695366016180&ref_url=https%253A%252F%252Fwww.google.com%252F

    https://www.ti.com/lit/an/sboa441/sboa441.pdf?ts=1695366046746&ref_url=https%253A%252F%252Fwww.google.com%252F

    The following application note reference is using parallel op amp to increase the higher output constant current. 

    https://www.ti.com/lit/an/sboa524/sboa524.pdf?ts=1695366121928&ref_url=https%253A%252F%252Fwww.google.com%252F

    I want to get the OPA Power consumption under the ioad current is constant 10mA(the current donest change),in this case ,the equation Popa=Vcc_diff * Icc + Vee_diff*Iee + Iq*(Vcc-Vee)-Iload*Vout  is right?and the Icc or Iee can get only with simulate or real test? Can I get it through other ways。

    If you are only interested the Iout = 10mA in the Howland current pump state, then you can use DC method to calculate the op amp's power dissipation.

    Since the OPA462 is using single supply rail, Vee = 0 ,so is Iee = 0. 

    Iout (constant) = 10mA  --> power dissipation is (120 - (10kΩ + 500Ω)*10mA))*10mA = (120-105)*10mA = 0.150W or 150mW.

    Iq = 3.2mA --> 120*3.7mA = 0.444 W or 444 mW (worst case)

    OPA462's total power dissipation when sourcing 10mA is: 150mW + 444mW = 0.594 W or 594mW 

    OPA462 Howland 10 mA Ipump 09222023.TSC

    If you are able to measure the OPA462's thermal pad's temperature, then you may be able to back calculate the IC's power dissipation under the condition. You have to do some modification in order to estimate the Tj's junction temperature. If you are able to measure Tj fairly accurately, then you will be able to estimate the OPA462's power dissipation. They are directly proportional to each other. 

    If you have other questions, please let me know. 

    Best,

    Raymond

  • Hi,Ray

    Thank you for replying to my message so late. It's 3pm in my city of Beijing.I will carefully read the message you gave me,

    BR

    Aape

  • Hi Wenhe,

     If you have other questions, please let me know. 

    Best,

    Raymond

  • Hi Wenhe, 

    I am going to close this inquiry. If you have other questions, please let me know. 

    Best,

    Raymond

  • Hi RAY,

    I'm currently designing the other parts of the circuit. Once I have completed all the designs, please help me check it again. The case can be temporarily closed

    BR

    Aape

  • Hi Wenhe, 

    If you encountered any issues in the design, please let me know. Also, there are other ways to design the constant current sources.

    Best,

    Raymond