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.

LMH6702QML-SP: LMH6702 Maximum Input Voltage

Part Number: LMH6702QML-SP
Other Parts Discussed in Thread: LMH6702

Good afternoon,

I am using this amplifier in my front-end circuitry for a lidar system.  I am AC coupling to the LMH6702, and supplying a +/- 5V rail.  Typically, we are feeding 15nS pulses that are a maximum of 900mV.  On the instrument however, we are having an EMI/EMC susceptibility relating to the detector.  From the data I've provided, can someone tell me what the maximum input pulse height can safely be for this part?  I haven't been able to find any information about that in the datasheets.

As always, thanks for your help!

Dave Durachka

  • HI Dave,

    Yes, I agree with you there is nothing in the datasheet. Checking our quality resource to understand if we have any information. Will get back to you tomorrow or Monday on this.

    Thanks
    Christian
  • The datasheet does provide an absolute maximum spec for "ESD Tolerance" of 1000V (human body model, 100pF, 1.5kohms). This is the maximum pulse height the device has survived under the defined test condition of human body model.

    thanks
    Christian
  • Careful, the ESD ratings are with the pins floating in space where specific ESD waveforms are zapped between all pin combination to ensure the steering diodes on pins take the energy into the absorption device and the device shows no DC parametric shifts before and after - nothing to do with in situ overdrive, 

    The LMH6702 is a current feedback design with an open loop buffer across the inputs specifying the V+ input can go RR input. not really with linear performance. But, the V- node will follow the V+ until it saturates. That happens 3.8V below the supply? Once it saturates then back bias voltages build across up to 3.8V total if the V+ input goes the supply. The concern is always Vebo's a transistor. 

    In the ABSmax is says the V+ can go RRinput, I guess implying nothing will break down - maybe, 

    If you need a simple discussion of CFA blocks, I was just putting this old app note I wrote into an article reference, 

    http://www.ti.com/lit/an/snoa366b/snoa366b.pdf

  • Hello David,

    The inputs of the LMH6702 are protected for brief ESD events, but not against intput overdrive voltages. 

    Is the input applied to the inverting or non inverting input?  Can you use a series resistor to protect against excessive input current? 

    As a general guideline I would say make sure that no more than 10mA of current flows into the input pins, and do not exceed 2V of differential input voltage.  This is a difficult topic to cover, though since the nature of the over voltage condition will determine how much damage it could induce. 

    TI does have a portfolio of transient voltage suppression devices, but I am not familiar enough with the portfolio to make a recommendation. 

    Regards,

    Loren

  • Loren, that +/-2V max diff comes from back to back diodes across VFA inputs, this is a CFA where the PDS says in the absmax is +/-Vcc saying you can put the full supply voltage at the V+ input and the V- input will stop following due to current mirror saturation about 3.8V below the supply - the question then is if that overvoltage distributes across the V+ to V- devices in a way that does not Vebo a transistor.
  • Hi Michael,

    We did use back to back diodes to protect the inputs on the LMH6702.  The abs max voltage of +- Vcc is the common mode input range - i.e. both inputs are close to the same voltage.  Based on my experience with the part, I'd recommend not going over +- 2V of differential input voltage. 

    Regards,

    Loren

  • thanks Loren, good input, probably should be in the AbsMax table. 

  • Thanks for all the great discussion on this topic, I really appreciate the time.  I will keep this absolute maximum value in my notes for present and future use.