Boost Converter Output ripple Calculation and Practical Prob

[mengghee]

boost converter output ripple

hiye everyone,

from one of my previous post, i have attached a picture of from output ripple from an oscilloscope and it is rather odd. but somebody told me it was alright. however, while i was calculating yesterday. i noticed that the practical output ripple is in the order or 200mV, however in practical calculation, it is only 2.2mV

the picture can be obtain from my previous post from this link



the equation used by me is

C=(D*Iout)/(ΔV*f)

i have used a 2200uF capacitor and my duty cycle is 0.5 and output current is 0.5A
frequency is 50 KHz. i have tried to solve for delta V and i obtained 2.2mV
however in practical design. i get it in the order or 100mV. can anybody tell me please ? thank you

regards,
mengghee

 

[Miguel Gaspar]

boost converter calculations

adjust your output filter at leas 1 decade below the frecuency os switching

 

[mengghee]

boost converter ripple

what do you mean one decade below switching frequency ?

 

[krishmaniac]

capacitor output ripple current boost converter

If eaxtly, your observation and calculations are correct. But you have missed the fact that the capacitor is not ideal in practice. It has series resistance often termed ESR"equivalent series resistance". the pulse current that goes throught he capacitor causes the drop across this resistance and often this dominated the capacitor ripple.
How to solve this then?
example:
if you are using 2000pF cap. do not put single cap of 2000pF. Put ten 200 pF cap in parallel, this will reduce the resistance. Then you see close to what you calculated.
try this out and let us know.

 

[mengghee]

boost output ripple

i do realize that the esr exist. but even with the esr, i don't think that it would make such a difference.....will it ? the practical value is 100 times the calculated value. i cannnot go to the lab anymore. term time finish ( i am a student ) however would really like to know how to solve this/

regards,
mengghee

 

[max0412]

calculate output ripple

Your equation doesn’t consider the primary contributor to ripple ESR. Check the data sheet on your cap for ESR value @ 50 kHz. Then solve for V pp out.

Ripple caused by output current:

Vpp = Iout/Co * ((Vout – Vin )/ (Vout * f))


Ripple caused by ESR:

Vppesr =* IpkL *ESR

* IpkL = (Peak inductor current)

For a converter in DCM IpkL can be high.

Vpp(total)=Vppesr + Vpp

Yes ESR can make a big difference.

 

[krishmaniac]

boost converter current ripple calculation

max ur 100% right. ESR makes a lot difference. Go and measure the ESR. parallel the cap then u will see how the ripple really reduce. Grab any SMPS design book.

 

[Vamsi Mocherla]

output ripple on a boost converter

I agree with max on the issues, I would like to add a couple more redundant points:

1. Check the ESR of the capacitor
2. Recalculate the sizes of the inductors and capacitors. The sizing of the inductor should be based on the ripple current. The sizing of the capacitor should be such that the output voltage ripple is very less.
3. Choose ceramic caps, if you can get them for the sizes, because they have negligible ESR
4. Did you put filter capacitors on the input because the input having a lot of ripple will spoil the buck driving transistor voltages. Use a high frequency capacitor bank there.

I have just summed up things for convenience. Hope that it helps

 

[eng_mohamed]

ripple boost converter

thanks u very much
with my best

 

[tekno1]

design boost converter with esr

mennghee,

"i do realize that the esr exist. but even with the esr, i don't think that it would make such a difference.....will it ? the practical value is 100 times the calculated value."

As people are saying ESR is the main culprid and in your calculations you do not considered it. Since it is the dominating ripple source it could be well 100 times of unrelated calculation value.

If you could not go to lab still trying to calculate ripple with capacitor ESR considered would give you much better fitting to your earlier bench measurements.

good luck ...