I'm reviewing a set of inductors for the Experimenting with Magnetic Components design challenge. Because it's a design challenge, I'd like to start with a working product. A switch mode DC converter. It's one of the standard circuits: the boost converter. A design that increases a DC voltage.
In post 1, I checked the schematics and the inductor. Here I'm building and testing the design.
|
Building the Boost Converter on a Proto Board
With the shopping basket I won in a Project14 contest, I bought a set of copper clad PCBs.
I used a box cutter to carve islands.
The transistor and diode come from my limited pile of recycle parts.
Instead of clipping the component leads, I used them as the connection and probe points.
I used a standard solder iron (no station) and leaded solder.
I used an oscilloscope, but in essence you only need a DC 9 - 12 V power supply, a function generator that can generate block waves with more then 8 Vpp, and a voltmeter.
Check jc2048's comment on a trap for young players when building a boost converter without feedback. I mitigated it by having the load always connected (it's soldered on the board).
Components:
L: Bourns RLB Series 1mH Radial Lead InductorBourns RLB Series 1mH Radial Lead Inductor
CIN: 120 µF 25 V
RGATE: 3.6 K
Q: 2SK2632LS N-CHannel MOSFET
D: SB360 SCHOTTKY BARRIER DIODE
COUT: 4.7 µF 450 V
RL: 2.7 K 2 W (this resistor dissipates up to 1.33 W when VOUT = 60 V)
I reviewed the inductor (part #6 of the kit for this design challenge) in the previous post.
MOSFET 2SK2632LS2SK2632LS
The MOSFET package is isolated. The heatsink can safely touch the PCB ground plane.
It's overkill for this circuit (like all other components, it's what I had lying around).
It is designed for switching, is fast and has a low RON.
I'm driving the transistor with VGS = 18 V.
VDS(max) = VOUT(max) = 60 V
ID(max) = VOUT(max) / RL = 60 V / 2.7 K = 22 mA
Schottky Diode SB360SB360
Again overkill. With the components I put in this experiment, and a larger capacitor, you could generate higher current.
First Test
I've tested the board, with 9 V input, 72 KHz 18 VPP block signal with 80% duty cycle.
The output was 32 V.
I'll do some more measurements in part 3, related to ripple.
