Boost converter not working with any inductor

Question

So I have been trying to design and build a boost converter which steps 5 V up to 12 V. More specifically, the output is to drive a fan which runs on 12 V, 70 mA (nominally).

As could also be seen in the image, the converter has an n-channel MOSFET, with the gate connected to a square wave oscillator that runs on 20 kHz with a 50% duty cycle and the source connected to ground.

The input is a 5 V-DC power source, with the negative side grounded and a 40 mH inductor between the positive terminal and the MOSFET's drain.

Also connected to the drain, is the anode of a diode (normally a Schotky diode is preferred for a boost converter, but I had none on hand so I used a more ordinary diode. LTSpice simulations indicated no problem with this discrepancy), with the cathode being connected to a 470 μf capacitor and a 171 Ω resistor (to emulate the load) connected in parallel to ground.

Although this setup is effective in LTSpice simulations, the test build is not functioning properly. The square wave oscillator and the MOSFET are functioning (at least on their own), as do all of the other components.

No matter what load or capacitor, the output voltage only hovers around 1.8 V. Changing the frequency of the oscillator and especially changing the inductance has not tangible effect. In fact, even putting all inductors I have in series or in parallel does not seem to effect it. Not matter the configuration, no step up voltage can be measured.

I simply do not know what is wrong, be it the inductors, the oscillator, or the MOSFET. Admittedly, LTSpice did not have the exact models for OPAMPs (used for the oscillator) and MOSFETs as I have in real life, but analogous substitutes were used in the simulation.

Answer 1

This is a good circuit to learn about boost converters, but open loop, it isn't good for much. You can't control the output voltage well.

Your inductor is way too large. For learning, 1 mH would be a reasonable value, if you run at a much lower frequency. On a solderless breadboard, you want to run at a lower freq anyway.

Inductor charging equation at 1 kHz, 50% duty cycle:

I = V/L * t = 5/0.001 * 0.0005 = 2.5 A. Your inductor needs to have a saturation rating less than this.

With 100 ohms load, I get 11 volts. Be sure to have a load, open loop, the output will go to hundreds of volts without a load.

Other common mistakes are: wrong type of inductor (needs to be able to handle large currents, if it looks like a resistor, it probably won't work), gate voltage not high enough.

You used a signal diode, you may have already blown it. A standard 1N400X, series will work at low frequencies (20 kHz), work but not very well, you need a fast recovery diode like a 1N4935.

If you built this on a solderless breadboard, be aware that getting it work can be difficult, the contact resistance is high. Think: will my circuit work with a 1 ohm resistor at every connection.

To learn more about how these work, you need to see the current. Since a home lab is unlikely to have a current probe, you can use a shunt resistor instead. A 0.1 ohm from the source to ground is a good place. You can infer the other currents from this current.

Show us a picture if these ideas don't help you enough.