[SOLVED] 100hz ripple current at the output of LLC half bridge converter

[biswaIITH]

I am designing a 700W battery charger with boost PFC + half bridge LLC resonant converter.

LLC half bridge converter is digital controlled with a output voltage loop control.

We are seeing a ripple current of around 6-7 A when delivering 12 A dc at the output. You can see the attached pic

Any one experienced it before & solved it by tuning PI loop of the LLC converter?

It would be great if you can share your experience here

 

[dick_freebird]

If your mains frequency is 50Hz then I'd guess that your problem lies
with the PFC / DC filter, like not enough filter C or a loop that can't
fully follow / suppress the line-bridge ripple before it overwhelms
the filtering you've got.

Check out some mid-points in the conversion lineup and maybe you
will find the culprit / soft spot.

 

[FvM]

Single phase PFC generates 100 Hz ripple by nature. To isolate it from load current, you need 1. sufficient DC link capacitance 2. a certain voltage regulation capability of the LLC.

 

[biswaIITH]

Single phase PFC generates 100 Hz ripple by nature. To isolate it from load current, you need 1. sufficient DC link capacitance 2. a certain voltage regulation capability of the LLC.

Ripple magnitude is way lesser when it is connected to electronic load. But magnitude is more when connected to any battery (Lead acid & Li-ion)

 

[Calm down]

I feel that it is PCB line interference. Please eliminate the issues of whether the ground wire of the control part, the ground wire of the power part, the current sampling point, the voltage sampling point, and the signal line pass by the strong interference source.

 

[KlausST]

Hi,

Ripple magnitude is way lesser when it is connected to electronic load. But magnitude is more when connected to any battery (Lead acid & Li-ion)

Electronic load mode:
* resistive?
* constant current?
* constant voltage?

Ripple magnitude:
* voltage?
* current?

As long as you talk about ripple current and resistive or constant current load, then what you experiece makes perfectly makes sense.

Klaus

 

[cupoftea]

Give detail of LLC....then we see what is going on.
Half bridge or full bridge?
C(resonant)
L(resonant)
L(mag)
NP/NS
Give Vin (av) and Vout (battery)

Is the frequency of your primary fet drive actually changing?
What is the average vin and the 100Hz peak to peak voltage on vin?

When you answer this, i will send you excel and ltspice sim of whats happening.

But in any case.....the ripple current you show probz doesnt matter...as long as the average is right.

 

[FvM]

It would be helpful to see circuit schematic with DC link capacitor values.
A single phase PFC controller suppresses 100 Hz ripple in no way. Question is if the LLC stage has voltage/current regulating capabilities and if the LLC controller makes use of it to achieve a clean output.

 

[Easy peasy]

Ripple current into a good quality battery can be high because the battery is very sensitive to volt ripple given its impedance at 100Hz is very low, i.e 10mV of ripple and 2 m-ohm battery Z = 5A ripple

this is why the PFC must have low ripple ( a lot of capacitance ) and the gain of the feed back loop on the LLC at 100Hz must be very high . .. basic physics and power electronics

 

[cupoftea]

yes...it would be interesting for you to see the battery datasheet, and see if that current ripple is actually damaging to the battery.
I too apart one offline 6kW charger and it had no electrolytic caps, or other big caps...so it was obviously giving high ripple current to the battery...presumably it didnt matter.

 

[Easy peasy]

battery don't care for charging only - however interestingly - Li-Fe-PO4 has a noticeable heating effect if charged by substantial 100Hz current - this is noted in papers and confirmed in our lab.