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.

LMH3401: LMH3401 Propagation delay

Part Number: LMH3401
Other Parts Discussed in Thread: TINA-TI, LM3401, TLV3801, TLV3604

Dear TI Expert

I Would like to know the propagation delay of LMH3401.  It is not listed in the datasheet

Actually I used LMH3401 for digital parts. (Single ended to Differential conversion)

So propagation delay is important to design circuit.

Could you check how much the delay is in case of 1 GHz input digital single ended data ?

Best Regrads

Sungsik Jung

  • Hello Sungsik,

      This will depend on several factors: your input signal characteristics (amplitude, rise-time or sine frequency), supplies, and output loading (especially capacitive). Gain would be another factor, but this is fixed for the LMH3401. 

      We do have a Tina-TI Reference Circuit and a PSpice Model of the LMH3401 if you would like to simulate the delay with your external components. It is located at this link: https://www.ti.com/product/LMH3401#design-tools-simulation under the Design & Development --> Design Tools & Simulation tab. The LMH3401 datasheet also has figures of the pulse response against various Vo as a measurement reference: Figure 36 and Figure 38. 

                

    Thank You,

    Sima 

  • Hello Sima

    Thanks for your quick updates.

    Actually, I did a simulation with PSpice and the result was  about 100 ~150 ps. (Conditions : 1 ns pulse , rising / falling 0.01ns , duty 50%)

    Is this simulation result a realistic(reasonable) value?

    I would like to use this part for high speed parallel signaling.

    So, I need to reduce the deviation between signals.

    That's why I ask for propagation delay.

    Could you give me the maximum and minimum values of the delay in LMH3401.

    Best Regards

    Sungsik Jung

  • Hi Sungsik,

    the LMH3401 is a fully analog amplifier normally not used for digital signalling. Because of that there's no propagation delay time specified in the datasheet. The speed performance of an analog amplifier is expressed in terms of "small-signal bandwidth" and "large-signal bandwidth" in the frequency domain and "slew rate", "rise time", "fall time" and "settling time" in the time domain.

    And there is another important point: Digital chips are usually driven into hard saturation at the input and they swing at the output also from hard saturation to hard saturation. An analog amplifier is totally different and is designed to run in a limited linear operating range and not in saturation. And when you do this with an analog amplifier you may suffer from a huge overdrive recovery time. An analog amplifier hanging in saturation appears to be much slower than when running in its linear operating range. Because of that the "overdrive recovery time" is another important specified parameter in the datasheet of an analog amplifier. See also figure 44 of datasheet.

    By the way, when you do the simulations don't forget to add some realistic load capacitance at the output of LM3401.

    The time delay between input and output signal of an analoge amplifier can be roughly estimated from the phase delay in the frequence response, provided you stay in the linear operating range. Calculate the time delay from the phase delay for 1, 2, 5 and 10GHz in picoseconds. This gives you an idea of what time delay has to be expected for a square wave input signal.  

    Kai

  • For digital signals, a comparator would be more suitable, and has a propagation delay specification.

    Fast comparators (e.g., TLV3604/5, TLV3801) have LVDS outputs. Would this work in your application?

  • Hello Sungsik,

       I agree with Kai and Clemens, are you saturating the amplifier due to the digital application (logic 1, 0)? If you are, a comparator would be a better option. 

    Thank you,

    Sima