Determining Value of Capacitor Between PV Panel and DC-DC Convertor.

[JohnDave]

Hi, I Am working on a Solar based DC-DC Converter its a BUCK-Boost converter, and As I read some of the articles related to the same topic Capacitor between PV Panel and DC-DC Converter is Mandatory.
So, anybody can guide to calculate the value for This Capacitor such as what is the proper formula or methods.
Thank you

 

[d123]

Hi,

Interesting to see articles you refer to. That's a big topic, worthy of study itself. Not sure if will help, it's about LC filters at input to SMPS, since 'An SMPS presents a negative impedance to its source supply.'

Some to look at if you want:
Power Tips #3 Damping the Input Filter (parts 1 of 2 and 2 of 2)
snva801 Analysis and Design of Input Filters flr DC-DC Circuits
SLUP340 SMPS Compensation Made Easy (p.31 - p.32, Input Filter Stability)

There are lots of online tutorials and application notes about SMPS input and output filters that can help.

If you just want a capacitor at the input, look at RC time constants, permissible inrush current or di/dt, (linear) soft-start circuits... You need the capacitor to hold up the input voltage for sudden step loads, same as output capacitance, I can't remember offhand document names but 'calculate bulk (input) capacitance' should bring up some useful results and I'd expect a good few online calculators.

 

[FvM]

I would expect a bypass capacitor to be part of the DC/DC converter, so that input voltage ripple is reduced to an acceptable level without relying on external filter capacitors.

 

[Easy peasy]

If in doubt, put 1000 - 2200uF at 1.5 x max input volts on the input to the converter - hopefully it runs MPPT so that it's output will reduce along with the inicdent sunlight - else it will latch up in current limit ... as the Vin falls

 

[BradtheRad]

Although a capacitor helps to smooth input current, consider making a complete LC filter.
Say the most you can extract from your PV panel (at MPPT) is 1/2 A continually.

The buck-boost converts this level internally into waveforms of greater peak current.



C1 provides the greater current pulses for the converter.
L1 has the triangle waveforms, generating pulses of current to the 12V battery. These pulses are greater than the 1/2 A produced by the PV panel.

 

[Easy peasy]

Hello Brad, um, where do you get the 0.5 amps from ... ?

 

[BradtheRad]

Hello Brad, um, where do you get the 0.5 amps from ... ?

Just to pull a figure out of a hat. I don't know whether 10V 0.5A is a realistic MPPT. Of course the converter must be designed to operate on parameters suited to the real PV panel's capabilities.

 

[treez]

For the input cap.....theres no maximum....it can be as big as you like....but you mightnt want to pay too much for it...so may try to find the lowest value you can use.

- - - Updated - - -

What is the power level of this buckboost?

- - - Updated - - -

You coudl just use enough capacitance so that you take the brunt of the switching ripple out of the input current.
I mean, say if it was a 200W converter....you could get away with a 10uF film capacitor, save you using a low lifetime electrolytic.
(but it does depend on your panel voltage, and power, ie , what your smps input current will be.
Also, if you use an LC input filter...then you dont want the LC resonant frequency being the same as the SMPS switching frequency...you would want it much lower than that
Also, the corner frequency of your input filter should be at least 10 times less than the crossover frequency of the SMPS.
Also, the output impedance of the input filter should be at least ten times less than the input resistance of the SMPS at the crossover frequency of the smps
The “crossover frequency” pertains to how fast the smps is at responding to transients in the load. A “slow” smps would see its vout dip far down after a sudden high load got switched on to the output of the smps.
I guess that most people kind of plump for it on experience at first…and then increase the capacitance at the smps input terminals if they see oscillations which they think are caused by an input filter with too high impedance as “seen” from the smps input terminals.
What you have to watch out for with dcdc converters with mega high crossover frequency, is that oscillations seen on the output following a transient, might actually be due to a too-high output impedance of the input filter.

Maybe you have loads of room and can just shovel in a huge input capacitor bank and be done with it?


I am sure you know that the LC input filter should have a C as the most downstream component...so that the SMPS "sees" the filter as a low impedance. (ie looking back into the input filter from the smps terminals should be low impednace)

I think to make things easier on those very cloudy days though.... a big input capacitor is good......since the panels can trickle into it and top it up and the smps wont quickly drain it if its big.

- - - Updated - - -