Buck converter question

[martin1]

Hi

I'm in the process of building a buck converter for my wind generator but to calculate the size of the coil you need to know the maximum current ripple. I have no idea what the current ripple should be for battery charging.

[ghurd]

It it a single phase or 3-ph?

G-

[martin1]

It´s a 3-phase

[ghurd]

The battery doesn't care about ripple current.

Seems like 7% figures into there somewhere on a 3-ph.  7% of the current flowing at the time between max and min amps.  Not sure if that is important for a converter.

Seems like the converter's efficiency goes up if the designed ripple current is higher.  Probably depends on the design, but I think I would leave quite a bit of ripple for battery charging.

I hope you get a better answer than that!

G-

[martin1]

Thanks ghurd

I don't think the input ripple is important. A fairly large capacitor will take care of that. I'm more concerned about the output ripple from the converter coil.

[ghurd]

That's what I mean about the efficiency.

I think it usually goes up if the output ripple is higher.

Tighter tolerances mean lower efficiency.

G-

[Airstream]

Batteries do care, just some more than others.

My 10-year Telecom AGMs warranty spells out required ripple specs second only to float voltage range...

"Permissible continuous AC ripple voltage at the battery terminals shall not exceed 1.5% peak-to-peak of the float charge voltage and shall not exceed 4% peak-to-peak instantaneously at any time."

[Flux]

Yes gell cells and possibly AGM on float do have specified ripple current ratings because the ripple leads to gassing and eventual loss of electrolyte.

For flooded cells or even AGM used for charge/discharge this is not really an issue unless the ripple is enormous.

Now back to the real issue. You say this is 3 phase, by that I assume the dc supply is rectified 3 phase. I assume you are not using a 3 phase interleaved buck converter.

The ripple from the 3 phase bridge may influence the converter input capacitor slightly but it will be low in relation to the ripple from the converter so you might as well forget it.

The ripple to the battery will be that from the converter and if it is in continuous conduction it will be too small to worry any battery other than a gell on continuous float. You will need filter capacitors on the dc side of the converter and those combined with the converter choke and the high frequency will keep the battery happy.

If it is a 3 phase interleaved converter the ripple requirements on the input capacitor will be reduced drastically and the output will be very smooth.

Flux

[BigBreaker]

When I simulated my buck design I found a lot of ripple and it depended strongly on the switching frequency rather than the input.

The generic filter on a buck after the switching stage is an inductor.  In my design that inductor needed to be very large in order reduce the ripple current to an acceptable level.  That was due in part to a preference for lower frequency operation with IGBTs rather than FETS.

For my design it made sense to split the inductor in two and place a large capacitor in between that is tied to ground.  This is often called a Butterworth filter.  It really kills the ripple with a big, cheap capacitor.  Something to consider as the copper for a big inductor can get expensive.

[martin1]

I'm also thinking of having two inductors. The idea is not to have them in series but to use them one at a time. One of the inductors will have higher inductance but handle less amps and will be used in low wind. The other will have lower inductance but be able to carry more amps, up to 25A and used in high wind.

[domwild]

From: "Buck-Converter Design Demystified" by Donald Schelle and Jorge Castorena, Maxim Integrated Products.

"Inductor Selection

L = (Vinmax - Vout) * Vout/Vinmax * 1/fSW * 1/(LIR * Ioutmax)

where fSW is the switching frequency, and

LIR is the inductor-current ratio expressed as a percentage of Iout (e.g., for a 300-mAp-p ripple current with a 1-A output, LIR = 0.3A/1 A = 0.3 LIR)." (End of quote)

The max. amps the inductor is supposed to be built for is given on the next page. As I do not have a black-belt in electrickery I suggest you download this short document, as it seems to be the best doc. re bucks I have ever come across. I cannot comment on the effects of ripple re batteries or efficiency.

Regards,

[domwild]

Another doc. is:

"Guide to Selecting Inductors for Switching Regulators" by Chris Likely, Cooper Electronic Technologies

Good Luck!