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.

TPSF12C1-Q1: Active EMI Filter

Part Number: TPSF12C1-Q1

Hello,

I am trying to design active EMI filter using TPSF12C1-Q1. Regarding that, firstly I have designed a passive filter by following procedure from Biricha digital. That passive filter circuit is shown below- 

Now, I want to replace common mode capacitors by TPSF12C1-Q1 and hence reduce CM choke size as explained in TI datasheet of the mentioned part number. But I am having some doubts in active filter circuit diagram by Texas instruments which is shown below-

Can you help regarding the following points-

  • How to select/calculate sense capacitors, Injection capacitor which are replacing Y-capacitors(CM) value ?  (In datasheet, values are for the given typical application example)
  • How to calculate reduced value of CM chokes compared to passive filter case?
  • Why there are two CM chokes in active filter circuit? 

I have gone through the datasheet and QuickStart calculator excel sheet for TPSF12C1-Q1, but not completely sure how to use it for my design as sheet is protected.

Quick response would be appreciated.

Thanks,

Jasvant Singh

  • Hi Jasvant,

    Is it possible to split the CM choke into two smaller CM chokes here? This is the ideal configuration for this AEF circuit (otherwise we rely on the CM inductance of the AC grid supply).

    the sense caps are typically 680pF, as this gives sufficient attenuation of line-frequency voltages to the SENSE pins and ultimately to the INJ output. The inject cap is typically 4.7nF or 10nF, as it effectively replaces the Y-caps between the chokes. Note the Y-cap value to GND also relates to the touch/leakge current spec for the particular application.

    In terms of CM inductance reduction, this is set by the active amplification factor of the inject cap, which is effectively 25dB (or ~26) at the frequency that dictates the size of the chokes (100-300kHz being typical range of switching frequencies). Thus, the two CM chokes can be reduced by a factor of 5 each (for example). on other words, the effective multiplication of the inject cap at the frequencies of interest allows the CM inductance to go lower and still maintain a target LC corner frequency.

    Regards,

    Tim

  • Hi Tim,

    Thanks for the reply.

    However, I want more clarity especially regarding the typical application design example given in datasheet of TPSF12C1-Q1.

    My design is also somewhat similar to the given example. Only difference is that for my design both AC/DC and DC/DC stage frequency is fixed.

    Can you help regarding the following points-

    • How to calculate GAEF? [I have referred SLYT812, will the same formula work here? If yes, then GAEF is very low due to low Csense value (680 PF)]
    • What is the sequential procedure to design the AEF filter for the given design example? 
    • Is there any reference document to design the passive filter for the given typical application example?(all components Cx, Cy, LDM, LCM)

    Thanks,

    Jasvant

  • Hi Jasvant,

    The expression for Gaef is not published yet, but you can use the quickstart calculator to check its plot. The design procedure follows the steps and sequence in the quickstart calculator.

    In terms of DM and CM filter design, you can refer to my e-book www.ti.com/lit/slyy208 and article series here: http://www.how2power.com/other/EMI_Guide.php (parts 15 and 16 in particular).

    Regards,

    Tim