Hi, Basic test circuit to determine the affect of adding a buck converter to PMA system.
PMA RPM will be constant. Switching frequency is constant. Logic of point A controls gate oscillator turning MOSFET on and off. I still have to check data sheets to determine if cap and diode voltage multiplier is needed. Comments welcome.
Lets find renewable solutions to reduce energy cost
I started my design with something like the 2106, but I had trouble getting it to bootstrap into operation if the load voltage was 24v or higher. At least for the chip I was trying, the load voltage had to be below the supply voltage of the chip. Do you know if that's a limit of the 2106?
Have a look at half-bridge gate driver chips. They handle all the dirty work for you when using two N-Channel MOSFETS in the synchronous buck circuit.
This is from the datasheet of one of the driver chips I have used:
Floating channel designed for bootstrap operationFully operational to +600 VTolerant to negative transient voltage, dV/dt immuneGate drive supply range from 10 V to 20 VUndervoltage lockout3.3 V, 5 V, and 15 V input logic compatibleCross-conduction prevention logicprogramable deadtimeHigh-side output in phase with inputShutdown input turns off both channelsMatched propagation delay for both channels
The output drivers feature a high pulse current buffer stage designed for minimum driver cross-conduction. The floating channel can be used to drive an N-channel power MOSFET or IGBT in the high-side configuration which operates from 10 V to 600 V.
Have a look at some of the data sheets to see how they hook up.
N MOSFETs are a lot cheaper and can have a much lower Rdson than a P MOSFET.
Have a look at Tim Nolan's schematic:
Don't connect the driver chip Vcc to the buck input though. My buck input can get up to 70Vdc when the wind is upto about 14 m/s which would destroy the driver chip. Most driver chips can only handle up to 25v for Vcc so I just used the 12vdc supply on the controller board. Note that Vcc supplies the +10vdc offset to the High side N-MOSFET gate drive through the boot strap circuit connected to the driver chip ( See Dboot and Cboot in the datasheet diagrams of the driver chips).
GoVertical wrote:Is there a reason why a N channel can not replace the P channel fet in the Buck circuit?Comments welcome?
Is there a reason why a N channel can not replace the P channel fet in the Buck circuit?
I think you can do that, but the problem with all the circuits is that gate drive voltage. Making it high enough, without violating VGS max, which is usually less than 20v.
Hi, the formulas assume ideal conditions, I should start looking for a compactor circuit to stops the control signal at low input voltage. Limiting inductor coil saturation and having a working control signal are the major challenges. It is nice to have options, thank you for the help I am receiving.
For a couple of years I have been using Tim Nolan's Solar MPPT circuit to do wind MPPT on one of my turbines. Tim has a really nice write up about the circuit on his site that you should check out:
I did modify his code to fit my needs but the basic algorithm stayed the same. I also had to beef up the FETs since I was working at a higher voltage when at full power. I wrote about it a bit on fieldlines:
Caleb wrote:Could something like this be used in that application?http://cds.linear.com/docs/Datasheet/3845fd.pdf
Could something like this be used in that application?
That's a nice part for a power supply, but I wonder if it could be used for this purpose. I guess you would set the output to 14v and hook a resistor between that and the battery?
What worries me is it's a power supply. Orignal idea was a fixed dutycycle fixed frequency device, which would basically just cut the input voltage down by N and multiply the current by N. I think this thing might try to compensate for a lower input voltage by pulling more current, which is the opposite of what you want to do with a turbine that's running low on wind.
I made my inductor by wrapping 120 turns of #18 around a soda can. (Then removing the can.) I believe it's in the area of 1 mH. I wanted more inductance so the current was more steady so I could measure it. It seems like dI of 2.8 amps is a lot for a 10 amp output.
Going into a battery you don't care about ripple and you won't need an output cap.