• State TI Thinks Resolved
  • Locked Locked
  • Replies 3 replies
  • Answers 1 answer
  • Subscribers 50 subscribers
  • Views 1161 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

OPA695: What is the maximum output voltage when driving a DC coupled 50 ohm load?

John Kolchmeyer
Prodigy 60 points
Part Number: OPA695

All of the examples in the OPA695 datasheet show the device driving a 100 ohm load. When driving a 50 ohm load, the exanples put a series 50 ohm resistor at the output of the OPA695. We need to send a 40nS linear ramp every 16uS with the maximum voltage swing allowed by the OPA695. Simulations show that I can get a visibly linear response to a 40nS ramp with a 6.2 volt peak to peak output voltage into 50 ohms. The output current is about +/- 3.1V/50ohm = 62mA which is within the 85C spec of +/-70mA. The amplifier output directly drives a 6" micro coax cable which has a 50 ohm termination to ground at the end. Will this work? Will the amplifier oscillate without a series 50 ohm resistor?

Also, if I increase the input signal such that I approach 70mA at the output, I get some visible nonlinear response to the 40nS ramp which is acceptable to us. However, will the device be damaged? Is this effect modeled correctly in the simulation?

I am applying a +/- 5.4V power supply in the simulation. Will this burn out the device when driving a 50 ohm load? (The power supply spec is +/-6V max, +/-6.5 absolute max.)

The OPA695 is being directly driven  by a DAC with a differential +/-25mA current output. The feedback resistor is 500 ohms. The OPA gain is 166 ohms. We are in the design phase.

  • 0
    TI__Genius 14066 points
    Hi John,

    The goal of impedance matching is to prevent the signal from being reflected at the termination, which causes distortion. To do so, the input (transmission line) impedance must match the load (termination) impedance. This is important when the length of your signal trace is more than a quarter of your signal wavelength. Since the amplifier output is low impedance, the 50 ohm series resistor is added to give the signal a 50 ohm source impedance to match the 50 ohm load, but it could also be 100 ohm series resistance and a 100 ohm load. Here is a reference: www.ti.com/.../slla311.pdf. As you know however, this halves the voltage at the load. Since you need the full output swing value and your coax is relatively short, is it possible to use a larger resistor value to terminate to ground, such as 100 ohm?

    Best regards,

    Sean
  • 0 in reply to Sean Cashin
    Prodigy 60 points

    Hi Sean,

    Thanks for your reply. The 50 load that we are driving is inside an optical modulator which we cannot change. Therefore the 6" coax is 50 ohms impedance. If we add an additional 100 ohms to ground we would exceed the current drive capability of the OPA. Question: If we exceed the drive capability of the OPA695 continuously (by say 30%), will we damage the device? Will the OPA go unstable and oscillate? We are willing to tolerate a certain amount of non-linear behavior at the limits but would like the linear region to still work.

    Thanks

    John

  • 0 in reply to John Kolchmeyer
    TI__Genius 15975 points
    Hi John,

    While operating the OPA695 on +/-5.4V supply, it is important that the maximum junction temperature should not be allowed to exceed greater than 150'C regardless of the current drive capability. The maximum junction temperature will primarily determine whether the device will be damaged or not for long-term reliability and continuous operation.

    The operating junction temperature is given by Tj = Ta + ThetajA*Pd.

    where,
    The Pd or power dissipation in the device consists of 1. quiescent power dissipation when no-load is connected (Pq) and 2. dynamic power dissipation or additional power dissipated internally in the device while driving the load (Pload). The quiescent power (Pq) is simply the specified no-load supply current (14.1mA) times the total supply across the part (10.8V). The Pload for grounded resistive load of 50-ohms will be at the maximum when the output voltage is fixed at one half of either supply voltage (for bipolar supplies). Under this condition, Pload = (Vs/2)^2/RL where the RL includes the feedback resistor.

    Assuming worst case example and SOT23-6 pin package,
    Pd(max) = 10.8*14.1mA + [(5.4/2)^2]/(50||(500+166)) ~ 153mW + 157mW = 310mW
    So, the Tj(max) = 85'C + (0.31W * 150'C/W) = 131.5'C

    The Tj(max) in your scenario is less than the maximum allowed Tj(max) of 150'C. So, you should be good to operate without damaging the part for long-term reliability. I don't think the OPA695 will go unstable and oscillate if the drive capability is increased.

    The other thing is that I would not recommend to directly drive the 6" micro coax cable with the OPA695 output. Instead, you could add a 5 to 10-ohm series resistor at the OPA695 output before driving the cable for stability purposes. The un-terminated source impedance is possible for your application because the rise time is only 40nsec.

    Best Regards,
    Rohit