Learning about Inductors

[higginsdj]

So I created a simple boost circuit I found (see image) but started with a 3904 NPN before switching to the 2N2222.  I used a Signal Generator to create the 2v PWM signal and just used a 10k resistor as the load.  The circuit draws 82mA on the 2v supply rail.  I understand what the circuit is doing and how (the basics), but what I don't understand is
1.  The relationship between frequency and output voltage and
2.  The 'ringing' shown by the oscilloscope and why the ringing changes with frequency.
3.  why reducing the 10K load to a 1k load reduced the output down to 6.4v.

I've attached a few sample oscilloscope snapshots (chan 1 - input PWM, chan 2 - output from inductor, chan 3 output voltage).  I started with a 3904 NPN (30khz - 1Mhz scope outputs) then switched it to a 2222 for the last plot which gave the highest voltage output at 360khz.

[ledtester]

2.  The 'ringing' shown by the oscilloscope and why the ringing changes with frequency.

The ringing is characteristic of that of a LC tank circuit -- i.e. an inductor in parallel with a capacitor.

This video at 1:17 does a pretty good job of explaining how the ringing is generated:

Oscillators, the Basic Tank Circuit 1
https://youtu.be/fQ4yRVEzXQA?t=1m17s

Even though your circuit doesn't explicitly have a capacitor in parallel with your inductor there are other sources of capacitance such as:

- stray capacitance due to circuit layout (are you using a breadboard?)
- capacitance inherent to the inductor (parasitic capacitance)
- capacitance inherent to other components in the circuit (such as the transistor)

Here is a Falstad simulation of a LC-tank (with a damping resistor):

https://falstad.com/circuit/circuitjs.html

You can start the oscillations by momentarily closing and opening the switch (upper right corner of the schematic).

3.  why reducing the 10K load to a 1k load reduced the output down to 6.4v.

The circuit works by periodically charging up the 47uF capacitor. Between these periodic charge-ups the load extracts current from the capacitor which reduces its voltage. A smaller valued resistor drains energy from a a capacitor faster than a larger valued resistor and so the average voltage of the cap between charge-ups will be smaller.

Note that even though "10K" is a larger number than "1K", we normally would say that a "10K load" is a smaller load than a "1K load" because it draws less current.

For more info look up "smoothing capacitor".

Here is a Falstad simulation of smoothing capacitor being charged up by a fullwave bridge rectifier:


(Falstad fullwave rectifier with smoothing cap)

The graph on the bottom shows the voltage on the capacitor. Note how it decreases between charge-ups. The slope of the decrease depends on the load resistor.

[higginsdj]

2> Interesting - thanks for the links and demo - I now understand the concept - yes, its on a breadboard. 
So is the frequency of the ringing oscillations fixed (ie with no physical circuit changes, the capacitance/inductance doesn't change)?  If so then this would explain why there is fewer oscillations in the ringing when I increased the frequency of the PWM input?

3> Doh...  Yes, of course, I should have realised this.

Thanks for the detailed explanations.