Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

1kW Flyback converter for battery charging

Status
Not open for further replies.
Is there any single dirver IC that can provide high side and low side drive for same output PWM's, that can be used for a two-switch flyback converter.
 
Any of the usual dual mosfet drivers can be used.

Functionally its exactly the same as driving a half bridge or a full bridge.
The only difference being upper and lower mosfets are driven in phase so both mosfets turn on and off together instead of alternately.
 
Thanks treez and Warpspeed for your help.

Is a simple buck converter better to implement for a 1kW battery charging or a 2 mosfet flyback would be better ?
 

At 1kW, from 230Vac mains, (207V ave) for a 1/2 bridge topology the current is 10A in the fets, so a half bridge standard circuit would work well, (F.B = 5A per fet, 2.5A ave) as you can't have current mode in an half bridge you would need to feed forward the HVDC bus ripple to cancel it on the o/p if you want low 120Hz ripple. Also the o/p diodes only have to handle 21A or so, with a decent o/p choke (200V if you use a centre tap full bridge 2 diode rectifier)

For a flyback, the peak current in the "mosfet" (igbt) will be closer to 20A (and the Tx) and the output "diode" will have to handle 84A peak (for 21A ave), so you see there are advantages to other ways, the 1/2 bridge and full bridge are tried and true and proven to work...
 

450 volt input dc and 65 volt, 12 A output is required.

What are issues of using buck converter rather than flyback converter ?

Does the inductor at the output of buck converter can create problem in output current control in DCM mode ?
 

What are issues of using buck converter rather than flyback converter ?
Click to expand...

A buck converter will have extreme current peaks and troughs. The waveforms are more severe than a flyback's.

All components must be rated for these extremes. You can share the burden by interleaving two or more buck converters.

Simulation of twin interleaved buck converter, with your specs.

 

I'm not sure a non-isolated buck converter is intended, a forward is a buck, as is a full bridge, ...
 

Batteries can be considered as super capacitors with distinct charge discharge cycles (curves). Some ripple on the charging voltage will be of no consequence *unless* the peak duration and magnitude (means energy) is rather high. For some lead-acid batteries, some ripple can be desirable too.
 

What worries me is where this 450 volts dc is coming from.

And the safety aspects of how the output of a buck regulator might possibly rise to the full 450v with no output load connected, or a fault condition.

Pretty lethal stuff.
 

My buck converter (post #86) is only conceptual, of course. It carries a high voltage hazard to the batteries, since your power supply is hundreds of volts. It is possible for a spark to ignite hydrogen gas. For instance, when touching or disconnecting a clip from the battery terminal. Battery explosions have been known to happen.

Therefore I agree with the recommendations that you use a topology which provides isolation, and this implies a transformer.
 

BradtheRad said:
My buck converter (post #86) is only conceptual, of course. It carries a high voltage hazard to the batteries, since your power supply is hundreds of volts. It is possible for a spark to ignite hydrogen gas. For instance, when touching or disconnecting a clip from the battery terminal. Battery explosions have been known to happen.

Therefore I agree with the recommendations that you use a topology which provides isolation, and this implies a transformer.
Click to expand...

Isolation will not avoid sparking when connecting /disconnecting. Formation of hydrogen gas is a very real danger. Only solution is to first make the connection and then apply the power. But the current supply *appears* to be a design for charging multiple batteries to be charged simultaneously.

It is a recipe for disaster.
 

for non sparking connection you generally need a high value resistor for pre-charge, with a relay to short the resistor when the voltages across the resistor are < 10V say...

- - - Updated - - -

For 450V in and "48" VDC out, a half bridge will do the 1kW easily and is possibly the best choice.
 

Easy peasy said:
for non sparking connection you generally need a high value resistor for pre-charge, with a relay to short the resistor when the voltages across the resistor are < 10V say...
Click to expand...

Are you suggesting that if the battery voltage is (say) 20V (discharged lead acid battery of nominal 24V) and the charger voltage is (say) 29V (common for charging a 24V battery), there will be no sparking?

OR

Can you bring up the voltage of the battery by charging it through a high value resistor?
 

I'm saying that for voltages less than 10V unless your fault current rating is higher than 200A you can safely close a relay
 

I want to know how can the DC output voltage ripple be reduced less than 10mV.

Please suggest me some techniques. Waiting for help :)
 

Less than 10mV ripple is an unusual requirement for a battery charger.

But adding one or more LC filter stages to the output will reduce the ripple by any desired amount.
 
  • Like
Reactions: Ravi_H

    R

    Ravi_H

    Points: 2
    Helpful Answer Positive Rating
Warpspeed said:
Less than 10mV ripple is an unusual requirement for a battery charger.
Click to expand...

Sorry I forgot to mention the application Sir. I was trying to built a power supply using flyback converter. I tried L-C filter but the Peak to Peak ripple was more than 30mV.
 

The LC filter need to match the requirement.
Is this low frequency ripple at the switching frequency, or high narrow switching spikes ?

In the first case you need sufficint L and C to do the job.

If its very narrow spikes, you probably do not need much, but the inductor needs to have very low capacitance between turns, something like a single layer rod choke. And the capacitor a good high frequency type with short leads.
 
  • Like
Reactions: Ravi_H

    R

    Ravi_H

    Points: 2
    Helpful Answer Positive Rating
Warpspeed said:
The LC filter need to match the requirement.
Is this low frequency ripple at the switching frequency, or high narrow switching spikes ?
Click to expand...
Sir, The output load frequency is varying from 250 Hz to 40 kHz and the switching frequency is 100 kHz. I tried LC filter with the cut off freq from 50 to 100 Hz, but the settling time increases.

I'm trying to reduce the output DC ripple of the converter and ensure the output has settled to the target in less than 0.6 seconds.
 

If the switching frequency is 100Khz, than the ripple frequency will be at 100 Khz.

Cut off frequency should be more like 10Khz, not 100 Hz.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top