# What's wrong with my boost converter circuit?

When I first created my circuit, everything worked fine. I didn't use a logic level mosfet at the time but I managed to acquire over 10V at the output (DC motor) and the next morning it stopped working. Mosfet was overheating and voltage didn't increase. I exchanged the mosfet with the current logic level one in the following picture and managed to get a peak of 14V at the output.

Again, the next morning it no longer works. I have no idea why. Everything is soldered in the circuit except for the source to ground and the motor which is connected by jumper wires. There were moments when the source wasnt grounded and left floating, the jumper wires aren't properly connected so sometimes the load would disconnect. Could my mosfet keep getting destroyed either because (1) my load is disconnected briefly, (2) my source is left floating or (3) the disconnection of my circuit abruptly since it seems that whenever I get it working, I go to sleep and wake up to the circuit not functioning. My assumption that the mosfets get destroyed is because when I measure the DtoS (in circuit) using the diode mode of my MM its always on at 100mv even when I try to turn it off by shorting GtoD.

Measuring the current of my fet from source to ground its 400ma so it definitely gets hot quick.

Is there anything I could do to keep my circuit from constantly malfunctioning?

• Load is disconnected briefly - without any output regulation like in your schematic, then yes. – winny Feb 3 at 22:59
• The inductor datasheet you link to is in the uH, I suggest you calculated the ripple current of your inductor. You may end up saturating the inductor with that circuit. Unloaded the output voltage will rise until something fails. – sstobbe Feb 4 at 0:39
• With the output unloaded on an unregulated supply, the output voltage will rise, as pointed out. It will rise until it's limited by something, and that something is most likely the FET, which will be destroyed one way or another. I suggest either using a proper boost controller, or at least detecting when your output voltage is over limit and shutting down the PWM. – TimWescott Feb 4 at 2:34
• Your inductor is $L=330\:\mu\text{H}$ with a series resistance of $460\:\text{m}\Omega$. Core saturation is indicated around $900\:\text{mA}$ (or at about $300\:\mu\text{Wb}$.) You write that you measured $400\:\text{mA}$ and I calculate that your inductor current should peak out at $I_\text{peak}= \frac{D=50\%\:\cdot\: 5\:\text{V}}{f=19.4\:\text{kHz}\:\cdot \:L=330\:\mu\text{H}}\approx390\:\text{mA}$, if you are using a 50% duty cycle. So the calculations and your measurements seem consistent on that point. (I'm discounting the series resistance which would lower the current.) – jonk Feb 4 at 4:51
• However, the peak magnetic charge for the first cycle will be about $130\:\mu\text{Wb}$. The inductor, at first, will only barely discharge any of that (not enough voltage drop across the inductor for the short time you allow to discharge.) So the second pulse will drive it upward towards about $250\:\mu\text{Wb}$. Which is pretty close to its saturation limit. In just two pulses! So it is almost certain you will initially drive the inductor into saturation -- long before the capacitor charges up enough to remedy that problem. So the NFET and inductor will dissipate more. – jonk Feb 4 at 4:55