• TI thinks resolved

THS4541: higher harmonic in my circuit

Part Number: THS4541

When I use THS4541 to build a lowpass multiple-feedback (MFB) filter ,I have higher harmonic than normal.( Hi-Resolution Mode)



AS we see in normal mode:


  • Hi Andoon,

    It seems your output swing is higher in the first scope capture (NOW) compared to the second scope capture (Normal mode). If the output swing gets close to the supply rails, then there is a possibility of noticing higher harmonics at the output. The harmonics could also deviate from normal operation if the output common mode voltage of the THS4541 is close to either of its supplies rather than being close to mid-supply.

    Some additional questions to better understand the issue:

    1. Are the scope captures taken directly at the THS4541 MFB filter outputs? If not, then what is the actual output voltage swing at each of the scope screen shots?

    2. What is the supply voltage and input signal frequency the THS4541 circuit operates on?

    3. How are you generating the differential input voltage to your circuit? Are you using a balun/transformer for this?

    4. Are you using any external filter for filtering out the harmonics generated by the signal source?

    Best Regards,


  • In reply to Rohit Bhat:


    Thanks for your answer.

    As you see, the first scope is higher than the second. Because the signal is not  through the circuit (pic 1).I just want to say the signal which I use to test the THS4541 is fine.

    I have thought carefully about the questions which you post, I find out  the signal at test point T1  is fine,but the T2 is worse . 

    the circuit is low pass in 5.5MHz, the frequency of the signal  is 1KHz to 2.5MHz. 

    I can not find out which is the problem (Res and Cap in MFB or the THS4541 ) .

    Thanks a lot.


  • In reply to Andoon:

    Hi Andoon,

    Even though you have implemented a 5.5 MHz low pass circuit, most of the harmonics from 1kHz to 2.5MHz (upto HD3 for 2.5 MHz) will pass through at the THS4541 output without being filtered. What is your measured voltage swing at both the input and output of the THS4541 MFB?

    It also looks like you are probing only the single-ended test points T1 and T2 of the circuit. If you probe only the single-ended input and outputs of the THS4541, you are bound to get higher even-order harmonics. I would recommend you to probe the circuit using a differential probe, so that the even order harmonic cancels out.

    I would also recommend you to use a spectrum analyzer to measure harmonics instead of generating FFT from the probed signal on an oscilloscope. This is because the analyzer has better dynamic range/SFDR performance compared to the oscilloscope.

    Best Regards,