Audio amp classes as fast as possible!

in the previous two parts of this series we talked about how Dax Convert digital to analog and why once you've got an analog signal you still need an amplifier to actually power your headphones or speakers and in this third and final part we're going to be talking about what the differences between the three main types of amplifiers used in audio are and how they work brought to you by headphones. com if you like educational content like this and want to support it consider headphones. com for your next audio purchase and by with confidence thanks to headphones.

com 365-day return policy so there are a number of different classes of amplifiers but in audio we almost always see the same three Class A Class A B and class D so we're going to focus mainly on those three throughout this video we'll be keeping things simple and easy to understand but if you'd like a more in-depth read into any of the topics covered there are some great resources Linked In the description let's start with the component that makes up the core of any solid state amplifier the transistor this is an electrical component with three terminals that can be used to control the flow of large electrical currents using a smaller input current say you had a hose with high press water coming out when and opened you need a way of being able to control and adjust the flow of this water since you can't just push back against it with your puny weak human hands so you install a valve this device allows water to pass through unrestricted when open but when turned needing only a small amount of force to do so it can slow or completely stop the water the transistor is the electrical equivalent of a valve allowing you to control the amount of current flowing across two of its terminals The Collector and the emitter by varying the current to its third terminal the base let's look at a classic BJT npn transistor other trans transistors are available here we apply a constant DC voltage VCC to The Collector and current is allowed to flow from the collector to the emitter when the transistor is on if there is nothing flowing into the base then the transistor is off and it acts as an insulator and prevents current from flowing from The Collector to the emitter if you apply an input current to the base this opens the valve and allows a larger current to flow from the collector to the emitter the more you apply to the base the more current flows across the transistor and so by varying this current applied to the transistor base feeding it a small version of our input signal we can vary the much larger current flowing from The Collector to the emitter and in doing so get a larger Amplified version of our input signal but there is one more thing and this is where the differences between Class A Class A B and Class B amplifiers come in a transistor will actually only switch on once a certain input current to the base is achieved it doesn't start at zero meaning we need to make sure that the transistor base is always above that threshold or biased so that any changes to the input will automatically result in a change to the Amplified output rather than just for below that threshold and doing nothing this chart shows an example of the relationship between the input to the base versus the output from the emitter you can see that there's a coff area where nothing is actually done and then a linear portion where the current across the transistor increases proportional to the change in the base current if we just left the input signal to the base as it was then if we fed it a typical sinewave input signal like this then the signal would fall below the cfff area of the transistor and would not be correctly Amplified so we need to shift or bias the input to the base to ensure that the false signal can be Amplified correctly by connecting a speaker here the small input signal to the base which we've biased into the active region of the transistor controls how much current is allowed to flow through the transistor and the connected speaker thereby moving the speaker and producing sound this simple setup is a class A amplifier one transistor buy us sufficiently to allow it to drive the full input cycle of our signal but wait there's a problem with a Class A Design like this we've biased our transistor so that even when there is no input signal the static or quent current is still deep within the active region the transistor is still on and current is still flowing even though we are not actually feeding any signal to the base which generates a lot of heat and waste a lot of energy this is why Class A amplifiers run hot they are always active and any current not being used to drive the load or speaker itself is just converted into heat so how do we solve this we make a Class B amplifier a Class B amplifier uses at least two transistors Each of which handles a different portion of the signal this top one is an npn or negative positive negative BJT same as is what we saw in the previous class a example but the bottom one is a PNP or positive negative positive BJT and the key difference here is that where the npn transistor activates when you apply a positive signal to the base the PNP transistor activates when you apply a negative signal to the base this means that when we feed our small input signal the positive half activates and is Amplified by the npn transistor and then when it becomes negative the npn transistor turns off and instead the P&P transistor activates and amplifies the negative half because of the shared Duty and no longer needing to have your transistor to amplify the full cycle of the signal you don't need to bias it deep into the active region like we did with class A each of these transistors can be biased right up to the activation threshold allowing any positive or negative input signal to activate and be Amplified by one of these transistors but when no signal is present neither transistor is active making Class B much more efficient but wait there's a problem transistors aren't perfect they don't switch on instantaneously and their activation region is not an exact Turning Point meaning with a basic Class B set up like this there may be some slight inaccuracy or delay when one transistor turns off and hands over to the other and this creates Distortion in the output signal called zeroc Crossing Distortion which is why actual Class B amplifiers are very rarely used in audio instead most audio amplifiers are class AB amplifiers these are still similar to class B amplifiers but rather than having the bias for each transistor set right to the activation Point each transistor is biased a little bit further meaning that the top transistor will amplify the full positive cycle of the signal and a little bit of the negative whilst the bottom one amplifies the full negative cycle of the signal and a little bit of the positive both transistors are active around the zeroc crossing point eliminating The zeroc Crossing Distortion that would otherwise be present when you suddenly switched from one transistor to the other whilst still being much more efficient than a full Class A amplifier due to not having such an extremely high bias and constantly passing High current even when no input signal is actually present so Class A amplifiers can be designed more simply and are often sought after due to the arguably cleaner approach to amplification with no sudden changes from one transistor to another but they are inefficient and produce a lot of heat which in turn requires more thorough and expensive heat dissipation and power supply components additionally all of this extra heat and power consumption can make it more difficult to get really low Distortion with a full class a design class B amplifiers solve these problems but introduce the issue of zeroc Crossing Distortion and Class A B amplifiers aim to solve zeroc Crossing Distortion whilst avoiding the inefficiencies of a full Class A Design so what about class D well strap in class D amplifiers work in a completely different way to class A ab and B amplifiers they do not directly amplify the input waveform Itself by continuously varying the current passing through a transistor instead they create a series of high frequency full amplitude pulses and their output can only ever be on or off let's examine the setup for a typical halfbridge class D amplifier this has a triangle wave generator and the analog input signal both of which are fed into a comparator this comparator is a circuit or component that looks at which of the two inputs is higher than the other and this comparator feeds a switching output stage which is a transistor that can switch swi between fully on or fully off extremely fast hundreds of thousands of times per second at least if we overlay the triangle wave signal and our sine wave input signal being fed to the comparator we can get a better view of what's happening at any point where the input signal is higher than the signal from the triangle wave generator the comparator sees this and tells the switching output stage to switch on and at any point where the input signal is lower than the triangle wave the comparator tells the switching output stage to switch off this leaves us with a series of high frequency full amplitude pulses of varying width according to whether the signal was higher or lower than the triangle wave at that time this is called pulse width modulation and to get the intended Amplified analog signal from this the output of the class D amplifier is passed through a low pass filter which filters out the high frequency components from the signal think of this as restricting how fast the output is able to move up or down as a simple visual example of this effect this is a 1 khz square wave but it's only a square wave because it has lots of very high frequency components if we start to progressively filter out these high frequency components we see this 1 khz square wave transform into a 1 khz sine wave a class D amplifier filter will try to remove anything above 20 khz so that all we are left with is the waveform consisting of audible band content and no unwanted ultrasonic components if you've seen the Dax 101 video you'll be familiar with the Flappy Bird analogy R2 our Dax and Class A or AB amps control the output by telling it exactly what to do exactly what amplitude it should be at a given point in time whereas one bit Delta Sigma Dax and class D amps use a method best explained by Flappy Bird you want to follow this path through the pipes but you can't just draw a line with your finger and have the bird follow it you have to pulse pressing and holding your finger to provide a 100% go up instruction or letting go to provide a 100% go down instruction the more frequently or the longer that you press the faster the bird moves up and the longer or more frequently you let go the faster the bird moves down the exact same thing is happening with the pulses produced by a class D amplifier and the low pass filter effect for Flappy Bird is simply the fact that the bird does not instantaneously hit the top when you tap the screen there's a restriction on how fast meaning how high a frequency the bird can change direction so class D amps and Delta Sigma decks are playing Flappy Bird and r2r decks and Class A amps are playing Forza hopefully that gave you some insight into the key principles behind the different amplifier classes you'll encounter in audio but if you have any questions about amplifiers dacks headphones or gear then head over to the headphones. com Discord server or the headphones. com Forum where I and other Wiggly air enthusiasts will endeavor to help I'm golden sound you're watching the headphon show by headphones.