Tags
speaker, pulser, oscillator, impedance, resonant frequency
Difficulty Rating
2 on my scale, if you're starting from here. But if you already built the last project, then this hack is just a 1!
Purpose
A simple change to our previous circuit, the Metronome. This change will make the pulses faster.
Bill of Materials
We need the circuit from Project 02, assembled and working, plus these parts:
Label | Description | Image |
---|---|---|
R4 | Resistor, 100kΩ (100,000 ohms) | |
C1 | Capacitor, 1 nF (1 nanofarad) | or |
Your capacitor most likely looks like one of three styles shown... or it might not. Capacitors come in many shapes and colors. That's OK — it should still work OK, as long as its wire leads aren't too thick and will fit into the small breadboard holes.
Assembly
We need to change two parts, C1 and R4.
Success is when you hear a tone coming from the speaker, and its pitch changes as you adjust the potentiometer.
Schematic
Images
I show four sets of waveforms. Each has a different potentiometer setting. Why show so many? Something interesting is going on...
Potentiometer set to 100% (fully clockwise). The fastest pulse rate.
Click on the image to see a larger version.
How Does It Work
What is going on with these waveforms? What happened to the sharp, clean pulses from the previous circuits?
Except for node n3, all the waveforms look very distorted from what they were. But even node n3, at certain potentiometer settings (like 75%), looks weird!
Hmmm...
It turns out that the speaker is the culprit. In the earlier circuits, an LED was the load — and an LED is a (relatively) simple load. A speaker is a complex load — even a little speaker like the one we're using. I should explain the speaker, first, before I explain the funny waveforms.
Speakers (again)
I introduced the speaker in the main project. Speakers are complex. The kind we're using is an electro-magnetic-mechanical device. They convert energy in electrical form into magnetic, then mechanical, and finally sound.
electrical → magnetic → mechanical → sound
Each conversion by itself is complex. The whole chain of events is greatly so. But I'll only discuss the speaker at a simple level.
Speakers have several characteristic properties, but I'll focus on just two: impedance and resonant frequency.
Impedance
Resistors have a very predicable and constant resistance. Well, of course — they were manufactured to be that way. The current passing thru the resistor can vary over a wide range, but its resistance will stay constant. Even if the current changes quickly (like the pulses across R3 in our LED flasher), the resistance stays constant.
Capacitors are also predicable but not constant with changing signal frequencies. Their resistance to changing voltages is high at low frequencies. Their resistance decreases predictably as the frequency goes up. (This is called an inverse relationship). At a given frequency, a higher-valued capacitor has lower resistance than a lower-valued capacitor. Capacitors cannot pass DC (direct-current) — they have infinite resistance at DC or "zero hertz", or it approaches infinite as the frequency gets slower. Because a capacitor's resistance varies with frequency, it's wrong to call it resistance: it's more correct to call it impedance (both use ohms as the units of measure). Capacitors aren't the only electronic part that have varying impedance.
Speakers do not have a constant impedance. They were manufactured to be good at something else – making sound – not having a predictable impedance. At very low frequencies and low currents, a speaker can look somewhat predictable. In fact, its resistance will be close to its impedance rating. But as you speed up the frequency of the signal, its impedance changes greatly. For example, a typical speaker rated at "8 ohms", may have a resistance of 5 ohms, an impedance of 8 ohms (measured at 1000 hertz), and peak impedance of 20 ohms (at its resonant frequency, like 2200 hertz), and then a decreasing impedance at increasing frequencies. Note that another model of speaker can have different impedance values and resonant frequencies than my example.
Another way to explain... In a true resistance, the voltage and current are perfectly aligned in time. That is, a 1-volt pulse across a 1000-ohm resistor will create 1-milliampere current pulse — the voltage and current pulses will start and end at the same times, no matter if the pulse lasts 1 second, 1 millisecond, or 1 nanosecond. But when a part has impedance, then the voltage across, and current thru, the speaker can be shifted in time (phase shifted). A 1-volt pulse across our speaker will create some amount of current at some point in time, but it can build up and trail off gradually, and maybe fluctuate or resonate in between those times.
Resonant Frequency
Resonant frequency is a physical property that shows up in a lot of places: in mechanics, acoustics,... and electronics. When we think of resonant frequency we usually think of guitar strings, organ pipes, bells, other musical instruments,... and even drums.
A speaker is like a small drum (a bongo). It has a resonant frequency. If you tap on it just right, you might hear it. It affects its electronic behavior too. It is the frequency at the speaker's peak (highest) impedance. At this one frequency: the speaker's cone (diaphragm) will fluctuate the greatest; the speaker makes its loudest sound; the peak current will be at its lowest level; the energy conversion (electrical →...→ sound) will be at its best. So with the right circuit, you should be able to actually hear it yourself.
(Note: I am ignoring acoustics to keep things simple. In real products, speakers are mounted inside of cases, cabinets, etc. And that can really affect the sound.)
Driving the Speaker
Our little oscillator circuit is not a perfect source of voltage pulses to the speaker. First, the speaker's resonant frequency is showing up on the output waveform (see node n2, potentiometer at 41%). Our little circuit doesn't have enough damping factor to restrict the speaker from resonating or "ringing".
Furthermore, our little circuit's operation can be influenced by a complex load. The positive feedback signal is not a clean pulse, but has speaker's effects mixed in — see the waveform at node n6, especially for potentiometer setting 75%.
If you slowly adjust the potentiometer ("doing a frequency sweep"), you'll hear changes in the tone quality. The speaker might even have more than one resonant frequency, which you might hear if you sweep it slowly and high enough.
In the Metronome project, we drove the speaker with short pulses at a very slow rate – lots of time in between pulses. Any fluctuations in the output current (caused by the speaker's resonances) died out long before the next pulse came along. But in this circuit, the pulses are coming very quickly. The pulses are either adding to, or subtracting from, the speaker's own fluctuations. Whether it adds or subtracts depends on the time between pulses, that is, what frequency the circuits oscillates at, or where your potentiometer adjustment is set. When the pulse and speaker-fluctuation match up in a positive way, you get a more "ringing" sound.
Simple Mods
Some of the mods I described in Project 2 can be applied here too: increasing the adjustment range, and changing the tone quality.
Known Weaknesses and Their Fixes
The sound output is somewhat limited:
- Sound Quality - Even if you try the various mods you may think the sound is still too harsh or "buzzy". How can we make it sound more mellow?
- Loudness - Unless you're in a quiet room, it will be hard to hear the tones. How can we make it louder?
These are good topics for a future article. They really deserve their own Project. But one quick answer would be to "baffle" the loudspeaker: mount it in a cabinet or box, so the sound coming only from the front of loudspeaker escapes outside. If you don't have a box, try putting a cardboard tube or open-end of a jar over the speaker — that will demonstrate the effect.
← Previous Article | ↑ | Next Article → |
Home Page |
No comments:
Post a Comment