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.

TPS55332-Q1: Compensation values calculation and EMI filter design

Part Number: TPS55332-Q1
Other Parts Discussed in Thread: LM5121, TPIC74101

Hi

I am using TP55332 IC for one of my application. The DC-DC converter is designed to operate as BOOST converter which provides 18.5V regulated output voltage from the input voltage range of 8V to 18V.

I want to know to compensation circuit values, so that with proper layout and correct compensation values, stability will not get affected.

Details of my circuit:

The operating frequency of my DC-DC converter is 350kHz,

output voltage is 18.5V, and output current is 1A

Chosen Inductor is 22uH power inductor

and output capacitance  of 4.7uF x 3 ceramic capacitors,

Input Capacitor is 4.7uF x 2 ceramic capacitors.

So, what should be the compensation values of Resistor and capacitor, so that the circuit will be stable?

I tried calculating from the datasheet, but unable to do it. Can you tell us how to calculate the values?

Also, I see the switching power losses in this switching regulator is very high, can any adjustments made so as to reduce the switching losses?

And for the EMI filter calculations of boost converter, can we use the Application note "AN-2162 Simple Success With Conducted EMI From DC-DC Converters" of Texas Instruments? Kindly suggest.

Thanks and Regards

Pratima V.

  • Thank you for considering TI part. Can you follow the TPS55332 datasheet section 8.2.2.8 and get your compensation value selected? Regarding the EMI filter, AN-2162 addresses a buck converter although the principle of the EMI filter selection is generic. The TPS55332 is a boost and the input current is not a pulse but a largely dc with ramp superimposed. Therefore the input current does not have much ripple (ac component) as in a buck, so the required filter would be smaller.
  • Hi

    Thanks for providing information. I referred the section 8.2.2.8 which gives the compensation values. But the concern is, it is directing us to some of the formulas.

    Equation 16 and equation 17.

    The "R" resistor is unknown. What does "R" represent? Since it is unknown, I am unable to calculate the value.

    Secondly in my application I used ceramic capacitors at the output. Total 3 capacitor with total value of 14 uF. And the output inductor is 22uH.

    So, the ESR of output capacitors will be very low, less than 50 mohm. Is this going to cause the instability in my circuit due to high gain etc?

    Kindly comment.

  • "R" is the min load resistance, which is "R" at the maximum load. This is explained after Equation (10).  Regarding your ceramic capacitor selection, it should be okay.

    Best Regards

  • Thanks for your Input.

    Still I am little bit doubtful that if 1A of current is taken for the output voltage of 20V, the switching losses in IC are very high, around 0.7W when calculated using the formula given in datasheet, and the total loss in IC is 1W.

    Could you please tell, is this ok or do I need to reduce the switching frequency from 350kHz to some lower frequency?

  • Lower frequency can surely reduce the switching losses and you can do the similar calculation to asess the power loss at a different frequency. Note the lower the frequency may require you to increase the inductor and filter capacitors. On the other hand, the IC should be able to handle 1W power loss in it because it has a thermal pad underneath to help dissipate the heat. Please to refer to the thermal resistance of the IC in the EC table and evaluate your system thermal performance.
  • Thanks for your input. This is really helpful.

    I checked the graph, and based on that I went for 1W of power loss.

    Further I have few more queries with Boost design using this IC.

    As you know that the device can sustain continuous input voltages upto 40V. And it can withstand Transients upto 60V.

    So, if we operate the device above 40V, say 45V continuous, which is below the absolute maximum rating of the device, will the device be able to sustain voltage or do we need to cut the power when the voltage exceeds?

    Second thing, When the device is not operating and the input voltage exceeds the output voltage, the output capacitor will keep on charging until the output voltage becomes equal to input voltage. Means it will pass input voltage to the output side of converter.

    I want to know, is there any way of reducing the charged output voltage or avoiding the charging of capacitor as the diode is used at the output side of IC.

    In my application, where I used the power supply, the input voltage range (9V to 18V nominal, maximum 25V) exceeds the output voltage (20V) (for some of electrical test).

    I think keeping the IC off will be better option but there should be some way of avoiding the charging of capacitor.

    Kindly provide your inputs or suggestions.

  • Regarding the voltage rating, VIN=45V may be okay or not okay. You really need to also check the SW pin voltage and make sure it is below 60V. The SW pin usually sees some spike voltage on top of the Vou during switching.

    For your 2nd queston, you are right that boost converter has not mechanism to stop the voltage feeding if VIN>Vout. You need to have additional switch to block off the path from the input to output when the circut is not switching. Refer to the LM5121, which support the load disconnect function.

    In your case, as your VIN can be greater than Vout during operation, you should employ a buck-boost converter like the TPIC74100 or TPIC74101.

    Thanks,
    Youhao
  • Hi,

    Thanks a lot for you valuable suggestion, it is really helpful to me in understanding and gives a good learning experience. I will need your full support till testing.

    As per the recommendation given in the datasheet, SW pin should be kept below 52V.

    Also, as my input voltage will be below 45V, so the SW pin voltage will be maintained below 60V.

    As per your input, "The SW pin usually sees some spike voltage on top of the Vout during switching". So, could you please tell what will be level of voltage level spike as I have used a diode which has reverse breakdown voltage of 40V. Kindly provide your inputs.

    We have finalised the components, so can't go for the Buck boost converter now. Thanks for your input.

    But I will try to keep the Boost IC off when input voltage exceeds the output voltage.

    In my board, I have used one buck converter IC and one boost converter IC and both operating at different switching frequency. And there are two different EMI filters for both regulators. The concern is that both the EMI filter inductors and capacitors are close to each other as one end of both the inductors/capacitors are tied to Battery line or Supply line.

    So, could you please suggest how to work on this as I think, the LC filter frequency will change as both the EMI filters are close. Kindly provide your inputs. I tried the two ways. See the below images for the details.

    The two images show the different orienation of input EMI filter capacitors. Blue circle around the EMI filters are shown. The Red colour net show the Supply line (US_PROT net), yellow is ground.When the Red colour net is replaced by a big plane of same signal instead of traces, it appears as if filter frequency is changing. In the second image it is very clear that frequencies may change due to different L and C used for both filter. Kindly suggest the suitable way to route the signals so that both serve their actual purpose and do not change the frequency of the filter.

    Left Blue circle is of Boost and right side blue circle is of Buck EMI filter in both the images.

    Kindly reply.

    The schematic of Boost section is also shown. Kindly see and let us know your comments if any for the same. The R63 is changed to 360kohm.

    Thanks

  • I want to know about the answer for the above query.

    Thanks

  • Hi Pratima,

    It seems you opened another thread about the EMI filter orientation question. I will close this one and let's continue discussion with your new thread. Thanks.