The Augmented Transistor Amplifier

Ajouté : 2026-05-10

[00:00:00]Hello and welcome back. Today I want to continue talking about uncommon transistor amplifier topologies by looking at the augmented amplifier.

[00:00:09]Similar to the circuit we looked at last time, this also uses a transformer to achieve a form of lossless feedback, but this time the main utility is increased linearity and lower input impedance.

[00:00:22]So, if you're curious about this amplifier, then keep watching.

[00:00:27]So, the circuit starts life as a common base amplifier. Signal comes in through the emitter and out through the collector.

[00:00:34]And the core idea behind the augmentation refers to improving the design by taking the signal from the emitter and feeding it into the base.

[00:00:43]The signal gets amplified and inverted.

[00:00:46]The actual implementation looks just like the initial circuit, but with a transformer which has one winding in the base line and the other in the emitter line.

[00:00:56]In general, the transformer follows a turns ratio of 1 to L, where L is the base winding which is larger.

[00:01:05]Now, for the regular common base amplifier, most of its parameters, like the input impedance and voltage gain, are related to the internal emitter resistance.

[00:01:15]This parameter being emitter current dependent.

[00:01:18]So, both the input impedance as well as the voltage gain are critically linked to this parameter.

[00:01:25]Now, with the augmented amplifier, the formulas are more or less the same, it's just that the value used for the internal emitter resistance is different. I mean, there's no change inside of the transistor, the internal resistance is the same, of course, but the equations will be handled using an augmented version of this parameter, which is the default one divided by 1 plus the turns ratio.

[00:01:50]In other words, the input impedance is decreased by 1 plus L times and the gain increased by this amount compared to the standalone common base amplifier.

[00:02:01]So, we are getting an amplifier with higher gain and lower input impedance, but the transistor is still in the same operating point.

[00:02:12]For more information on this amplifier, I will again highlight the patent that describes it.

[00:02:19]So, this is quite an extensive analysis on the various implementations and the mathematics behind this amplifier, so it's quite nice read if you need more data.

[00:02:30]And well, both the patent as well as the paper I mentioned last time are written by Chris Trask.

[00:02:37]So, I do recommend both of these documents since these are some of the best descriptions of this type of amplifier that I could find.

[00:02:44]So, while the theory seems sound, we should be getting some interesting properties, let's also try verifying things in the circuit simulator.

[00:02:53]There are three main things to look at.

[00:02:56]The input impedance, the gain, and the linearity.

[00:03:00]Let's take them one at a time.

[00:03:02]For the first set of simulations, I prepared both the regular common base amplifier on the bottom side as well as the augmented amplifier on the top side.

[00:03:12]I used the same transistor model, resistors, input signal source, and load with the only difference between the two circuits being the addition of the transformer.

[00:03:24]So, this has a 4.6 microhenry inductance in the emitter circuit and on the base side I use the same value multiplied by the square of the turns ratio. So, this L is a parameter that we will be stepping through multiple values just to see how accurately the amplifier parameters change with the turns ratio.

[00:03:44]Finally, the simulation type is AC, so we're looking at how the behavior changes with frequency.

[00:03:50]And I also added in the.net statements to extract the network parameters for both circuits.

[00:03:57]So, if we run the circuit, we can look at the results one at a time. First, the input impedance. We can get this from the.net statement and for the classical amplifier, we expect this to be in the 5 ohm range.

[00:04:12]And well, if we check, we are getting very close to this value.

[00:04:18]There is a bit of variation, but it's more or less as expected.

[00:04:22]Now, if we turn to the augmented amplifier, well, we are expecting a smaller value by the turns ratio.

[00:04:31]So, the plot is showing us four different curves corresponding to the four values of turns ratio that were simulated. And well, first we can see that the impedance is not perfectly flat. There is a drop off on the lower and higher frequency side. So, this is to be expected because of the transformer, but for the exact value, at least in the linear bit, we are getting about 2.7, 1.8, and so on.

[00:04:58]But to complete this comparison, we don't really want to see the exact values, but rather the ratios between the augmented amplifier and the classical one.

[00:05:10]So, if we plot this out, well, we are getting some very nice values.

[00:05:16]We're getting a ratio of two, three, four, and five.

[00:05:22]So, almost identical to what the theory was telling us.

[00:05:27]Now, we can take a similar approach and check the gain as well.

[00:05:31]For the regular amplifier, we are expecting a gain equal to the load impedance of 50 ohms divided by the internal emitter resistance.

[00:05:40]So, it should be about 10.

[00:05:42]And if we check the actual value, we are getting very close to this, about 9.2.

[00:05:49]For the augmented amplifier, if we run the same setup, we are getting some larger values, about 18, 27, 36, and 45.

[00:06:02]Values which do seem to be multiples of our basic circuit.

[00:06:07]And sure enough, if we plot out the ratio, well, we are again getting some very nice numbers, almost two, almost three, four, and five.

[00:06:19]Again, we are seeing a very good correlation between theory and the simulation results. So, at least for the gain and the input impedance, we are getting almost identical values to the numerical predictions.

[00:06:33]For these two parameters at least, I was honestly shocked at how well the numbers lined up.

[00:06:39]Something must be wrong in the simulator.

[00:06:41]It can't be that close.

[00:06:44]Anyway, the last thing I wanted to look at is the linearity improvement. And this is where things go a bit fuzzy.

[00:06:52]The expectation is not all that clear and the result, while it's there, it's not as obvious as before.

[00:07:00]So, for this simulation, I turn to a transient type and I'm injecting a 2 MHz signal on the input. For the amplitude, I'm using a parameter that is being swept over multiple values and for the augmented amplifier, this value is also divided by the turns ratio.

[00:07:19]And on the regular amplifier side, I also multiply this by a factor of 1.1 just to get the same amplitude when the turns ratio is one. Finally, to make the linearity evaluation, I added a couple.four statements which will extract the harmonic distortion that is appearing.

[00:07:36]So, if we run the simulation, we can first check that the two output amplitudes are more or less the same and for the actual results, we need to turn to the output error log.

[00:07:47]So, here for the common base amplifier, we can see the distortion levels of 0.11, 1.2, and 7.8% whereas for the augmented amplifier, we are getting 0.05, 0.6, and 4.6%.

[00:08:06]So, roughly half.

[00:08:08]Now, we can also change the turns ratio.

[00:08:11]So, rather than a value of one, we can set it to four.

[00:08:15]This also requires the amplitude in the regular common base amplifier to be changed to 0.3. And well, if we resimulate, first we can observe that the output amplitudes are more or less the same again.

[00:08:27]And for the distortion, for the common base amplifier, since we're inputting a smaller input signal, the values have changed a bit, so the distortion is 0.03, 0.3, and 1% but for the augmented amplifier, we are getting 0.005, 0.06, and 0.18%.

[00:08:50]So, the difference in between the two is starting to get bigger.

[00:08:53]In other words, the augmented amplifier is getting lower values of distortion when outputting the same amplitude signal in the same load.

[00:09:03]So, while the simulator does show that the augmented amplifier is superior to the regular common base topology, of the three parameters we looked at, only the input impedance and the gain are the things that can be clearly highlighted. Therefore, I decided to only test those out on a practical circuit.

[00:09:22]For that, I built the amplifier we were just simulating and decided to probe it a bit.

[00:09:28]From a schematic point of view, it's more or less the same thing that we simulated. I added in a few capacitors which are not placed, but these don't really have that much of an impact. And for the transformer, I'm using a 4 to 1 turns ratio. So, the four is on the base side and the one on the emitter side.

[00:09:47]The actual transformer has about 20 and five turns.

[00:09:51]I used different color wire just to make things a bit more clear.

[00:09:56]So, to test it out, I connected the board to a power supply set to 7 volts.

[00:10:02]And then I have a 50 ohm signal generator connected to the input using a coax line.

[00:10:07]And finally, the two oscilloscope probes are placed one on the input and one on the output to observe the signals passing through.

[00:10:16]As load, I added a 50 ohm resistor directly on the board to the output terminals.

[00:10:21]If we now turn to the measurement equipment, we can first use the oscilloscope and wave generator combo to check the circuit is actually working.

[00:10:30]So, I set the generator to output a 1 MHz 200 mV sine wave. So, this should be well in the stable operating region of the amplifier.

[00:10:40]And if we check using the oscilloscope, well, the input, the yellow channel, is very, very small, something in the 5 mV amplitude range. This is clearly indicating the low input impedance value. And while the output on the blue channel has a much, much higher value.

[00:10:57]So, 180 mV approximately.

[00:11:01]So, the circuit does have quite a bit of gain.

[00:11:04]Now, we can, however, get a much better view of what's going on if we turn to the network analyzer. So, this is set to sweep between 50 kHz and 25 MHz.

[00:11:16]And well, if we let the measurement run, we can first have a look at only the first channel, and rather than expressing in decibels, we can express it in percentages. And here, we are seeing a value of around 3.8%.

[00:11:31]So, what this is telling us is that out of the initial signal amplitude, the input channel is seeing an impedance which is creating a voltage divider that reduces the value to this specific amount. So, if we run the numbers a bit, this is indicating an input impedance of around 1.97 ohms.

[00:11:51]It's not exactly the 1.07 ohms that we should be getting, but it's still a very low value. If we now turn to the second channel, and we now look at the decibels, well, we can see the gain of the amplifier is relatively flat from about 200 kHz up until around 10 MHz, and the value is about 27 decibels.

[00:12:14]Again, this is smaller than the 33 that we're supposed to be getting, but it's still a very high gain value.

[00:12:21]Now, in all fairness, in the simulator, we were using an ideal transformer. So, the smaller values that we're getting are probably related to the non-ideal coupling factor that the transformer is presenting. But regardless, the amplifier is working as expected. It is providing a very low input impedance, and it also has quite a lot of gain.

[00:12:44]The last thing to mention today is related to why having a lower emitter resistance could be a good thing.

[00:12:50]This is a parameter related to the emitter current. So, you either increase the current or use the augmented amplifier.

[00:12:58]While using a transformer makes things complicated, the alternative, having higher current, could also bring different issues.

[00:13:08]Now, this is the data sheet of the BFR92A, a wideband NPN transistor, which features low noise and low intermodulation distortion.

[00:13:19]And the reason why I'm bringing up this particular transistor is that the data sheet also contains some characteristic curves describing these two behaviors.

[00:13:29]So, first of all, looking at the noise figure, we can first observe that the noise figure is frequency dependent. Higher frequency generates more noise, and well, not much we can do about that.

[00:13:41]But the other thing we can observe is that the noise is collector current dependent.

[00:13:46]And here, we can do something.

[00:13:49]So, we can observe that using relatively low collector currents is better than higher collector currents.

[00:13:57]Now, another thing that we can observe, if we scroll down a bit, is the intermodulation distortion behavior. And here, we are seeing a slightly different story.

[00:14:07]The best performance, so the smallest distortion, is appearing at relatively high collector currents.

[00:14:16]So, in general, when you're designing the operating point of an amplifier, you need to choose between the two.

[00:14:22]What do you prioritize, low noise or low distortion?

[00:14:26]And well, this is an important thing when it comes to the augmented amplifier, since you can run it at relatively low current to get the noise benefit, and the augmentation provides negative feedback, which improves the distortion behavior.

[00:14:41]So, basically, you're getting the best of both worlds by using the augmentation technique.

[00:14:49]Now, while this is not a very common design, I did find this principle used in the well-good loop antenna as described on the M1GEO website. So, this type of loop antenna is considered by some to be one of the best loop antenna designs that you can make. And even though it's not obvious, the transformer uses the emitter to base feedback principle. So, both the base and emitter are found on the same transformer.

[00:15:14]Therefore, this setup should provide the same type of low input impedance as you get in the single transistor circuit.

[00:15:21]It's just that here, we have a more complex differential amplifier.

[00:15:26]In the end, the augmented amplifier does present some unique characteristics, which makes it an interesting choice when you need a low noise, low distortion, and low input impedance amplifier.

[00:15:37]Since the transformer only uses two windings, you can make this amplifier using commercially available parts.

[00:15:44]So, building such a circuit shouldn't be all that difficult.

[00:15:48]Now, whether you need it or not, as always, depends on your particular use case.

[00:15:54]So, with that said, hope you enjoyed this video. And if so, there are more similar videos on my channel that you might want to check out.

[00:16:00]And if you want to be updated with my latest releases, also consider subscribing. See you next time. Bye-bye.