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.

Questions about the bandpass filter of TPIC8101

Other Parts Discussed in Thread: TPIC8101

Hello,

My name is Chentao and I am using TPIC8101 for knock detection.  I have read the TPIC8101 manual many times, but there is not enough information about the bandpass filter. So I come here for help. I will be appricate if anyone can help me.

 1 First I want to confirm one thing, the Q-factor. See the following website:

John Griffith said that the Q-Factor is 2.3, but I have heared that the Q-Factor is fixed to 2.5.  So I want to confirm which one is right. 

Is there any method to measure the Q-Factor?

And why the Q-factor is fixed?    

2 The detail information about  the bandpass filter. FIR or IIR, the filter order,  and the type: Direct Form I or Direct Form II or other type?

3 About the coefficients of the bandpass filter.

 As far as we know, the center freqency of the filter can be configured from 1.22KHZ to 19.98KHZ, so there are maybe 64 coefficient sets in the TPIC8101.  Where are the coefficient sets stored? Does the TPIC8101 has its own ROM to store the coefficient sets?  

 Or the coefficients are calculated by the TPIC8101 due to the center frequency configuration?

4 About the transfer function of the TPIC8101. See the following picture: Because w = 2 * pi * f, so we should use Vin = VIN * sin(2 * pi * Fbp * t).  But in the TPIC8101 manual, it always use Vin = VIN * sin(pi * fbp * t), why? Is there something wrong?

5 About the calculation of the tc. Tc is typically  Tint / (2 * pi * Vout). But how do we get this formular?

That is all,  waiting for your reply and best regards

  • The appropriate Applications Engineer has been assigned to respond accordingly. Thank you for your patience.
  • Chenato,

    1. The Q factor is 2.3.  Where did you hear that the Q factor was 2.5?  The Q factor is calculated according to this formula:  Q = f_center/(f_c2 - f_c1)

    2. It is a second order IIR bandpass filter

    3. You can find the details about the bandpass filter equation at the top of page 18 on the TPIC8101 datasheet.

    4. I think that you're right.  When I was working through the transfer derivation by hand with one of my colleagues a while back, we were confused because we were getting an extra factor of two compared to the datasheet version; however, when you change it to sin(2*pi*f_BP*t), it works out.  :-)  I'll verify this, and correct the datasheet.  The final transfer function is correct and has been verified through testing many times.

    5. I'm not sure how this formula was derived.  I'll think on it and get back to you.

  • Hello Clancy:

         Thank you very much for your kindly reply, your answer helps me a lot. 

         I have just found the new version of the TPIC8101, and I found the bandpass filter equation at the top of page 18, that is great! 

         About the Q value:  I got an application note from one of our customs, the note said that the Q value is 2.5, see the picture below.

         But now I think maybe they made a mistake. I will use the Q value in the TPIC8101 datasheet, which is 2.3.

        

          I will waiting for your reply, and best wishes for you.

          Chentao