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.

THS3217: Harmonic distortions

Part Number: THS3217

Hi! 

Could you, please tell me, what i'm doing wrong?

There are graphs of HD2 and HD3 in the datasheet of ths3217 t. It's easy to see, that if a signal has amplitude about 2,5 V in the output of ths3217, there will be about -50 dBc HD3 at 60 MHz on 100 Ohm load. Simulation result is more worse: only about -42 dBc with output signal amplitude 0 dBV (1V amplitude).

 8371.THS3217_HD.TSC

  • Well Anton, 

    The sim models do not support HD performance -it may give you a result in sim, but not related to part performance. 

  • Hi Anton,

    allow the circuit to settle and choose a suited "Sampling start time":

    Kai

  • There are a number of simulation tricks to "improve" your results -the basic one is to start your transient as a cosine wave to get a 0 dV/dT at time zero - you sine wave sim has a lot of immediate dV/dT things to settle out to a good sim - this was basic for any IC designer where the full transistor models will often give good HD3 results, but usually not HD2 unless you can randomize internal mismatches. Here is what your stimulus looks like now, 

    If I change the phase on your sources to start out as a cosine, I get this output time waveform, I needed this to start at 1V but apparently the initial value solution is not doing that, so this is not quite right yet, 

    Anyway, going to 50Mhz and starting at 100nsec on a 200nsec sim, gives another different result - none of these are inside the model - just sim artifacts. 

    Here is the model header on what is modelled - you will not see harmonic distortion in this (or any other I suspect) amplifier model. 

  • Hi, Kai!

    Thank you for answering my question. I tried to use your advice and HD2 and HD 3 became much more better then datasheet's values... Where is the truth? In the datasheet? 

    Best wishes, 

    Anton

  • Hi Michael!

    Thank you for your detailed answer. In this case i will believe in datasheet's characteristics.

    Best wishes,

    Anton

  • Hi Anton,

    Michael is right. If the model does not model harmonic distortion, then running the "Fourier Analysis" is entirely senseless, of course. I apologize for giving wrong support.

    Nevertheless, the trick of choosing a suited "Sampling start time" is still valid:

    Thanks for the lesson, Michael :-)

    Kai