On a previous blog post, I’ve been experimenting trying to build a mic preamplifier incorporating vacuum tubes. If you missed it, check it out. http://www.fivefishaudio.com/blog/adventures-in-building-a-tube-based-mic-preamp/
I discovered a few issues with our prototype tube preamp circuit. First, it’s the low voltage gain. We’re only able to amplify 7 times the input voltage, or approximately 17dB gain out of this medium-mu tube.
The second issue is the maximum output signal level we achieved is only +15dBu.
The third issue is the output voltage collapsed when I tried hooking up a powered monitor speaker to the output of our basic tube amplifier. We went from a nice 4.something Volts rms output, to less than half of that… and not only that, the waveform looked real bad. As I said earlier, operating this tube at a low voltage at low current (Volts x Current) = low power.
The low gain isn’t particularly a problem. It just means we need to add more gain stages to our design. Clearly, a single-stage isn’t going to cut it. And by adding a few more stages downstream, I’m sure we can also take care of increasing the maximum output level capability.
On the third issue, i.e. preventing loading of our vacuum tube amplifier, we need to match impedances, and/or make our output stage beefier so it can handle whatever we connect to it.
For curiosity’s sake, let’s find out the output impedance of our vacuum tube stage. And instead of going theoretical, let’s actually measure some voltages and do some simple calculations.
But how do you measure the output impedance of an amplifier stage (for that matter, any amplifier unit, equipment or module)?
Any amplifier output can be thought of as an “ideal” resistor in series with an “ideal” voltage source. Like this:
The “DEVICE UNDER TEST” in the diagram above can be a single amplifier stage in a circuit, or a finished product, like a preamp module, or a function generator, or some other equipment.
Ro is the “Output Impedance” of the amplifier stage. To calculate Ro is really simple.
We’ll feed our amplifier stage some signal, in this case I chose a typical 1Khz sine wave signal, and measure the output voltage unloaded.
Then we connect a *known* resistor value across the outputs, and measure again the output voltage, i.e. measuring across the load resistor. As you can see, the load resistor is now in series with Ro, and the 2 resistors basically forms a series voltage divider.
In our vacuum tube stage, I measured an unloaded voltage output of 3.68Vpp.
Putting a 10K resistor across the output terminal of the tube stage, the measured output voltage dropped to 1Vpp.
Now we can solve for Ro using the formula:
Ro = RL x (Vo/VL – 1)
; where Vo = unloaded output voltage
; VL = loaded output voltage
; RL = load resistor
Ro = 10,000 x (3.68/1.00 – 1)
Ro = 26,800 ohms
Using our measured values above, we get an Ro of 26,800 ohms (or 26K8). As you can see, it’s a very high impedance output, and no wonder our voltage collapsed when we hooked this tube stage directly to the inputs of a powered monitor speaker.
ASIDE: I can’t find specs for the input impedance of my Roland MA-8 speakers, but being a typical line level device, it’s probably 10K input impedance. Normally, 10K input is considered high input impedance…. but to our vacuum tube stage which has an output impedance of 28.8K, this 10K input impedance looks like a low impedance (relatively speaking).
Ideally, we want the output impedance to be low. So how can we fix this problem?
The first thought that came to mind is I need to add an output buffer stage after the vacuum tube stage. This buffer will present an even higher impedance for the tube output stage, and it will have a low output impedance for the next stage. There’s many ways to create a buffer circuit, using a JFET transistor amp for example. But for this particular case, I need something quick to confirm if my proposed solution is correct.
I grabbed one of my FAB MODULES. These are basically “Lego” building blocks for circuits that I created a few years back. It’s very useful for quick prototyping, or for incorporation into your DIY project. All you need to do is interconnect them together. I have several modules designed, each module serving a unique purpose. In fact, one can quickly build and experiment with different preamp configurations using these FAB Modules. Like so:
Okay, back to the topic… For the output buffer that we want, I picked a FAB4300 module. It’s basically a Unity Buffer, that we’ll immediately put after the Vacuum Tube output stage.
Now, I think an OpAmp is perfect for this buffering task.
An “ideal” OpAmp has an infinite input impedance — our vacuum tube stage would like that! No more loading. An “ideal” OpAmp also has a zero output impedance, Ro. The next stage or device would like that! We can deliver all our output voltage to whatever following stage with none or minimal loss. It’s also inexpensive, and the circuit is very easy to implement.
And… we can configure the OpAmp to give us some additional voltage gain, acting like another “boost” stage. This basically solves all the (3) issues I outlined at the start of this article!
So I wired the output of the vacuum stage to our FAB4300 Unity Buffer module, and repeated the same voltage measurements, both unloaded and loaded to measure our new Output Impedance.
The result? It’s perfect! No voltage loss!
Using the same formula for Ro above, let’s recompute the Output Impedance.
Ro = RL x (Vo/VL – 1)
Ro = 10,000 x (3.68/3.68 – 1)
Ro = 0 ohms
WAIT!… Is that even correct? We have zero ohms output impedance? If we’re living in an “ideal” world with unicorns and fairies then I would believe that. Did we screw something up? How can we get a 0-ohm answer?
Now, you have to remember, I’m just using the built-in display measurements on my scope, which is limited to only 2 decimal places. But for sanity check, let’s consult the manufacturer’s datasheet for our OpAmp.
Well, 0.01 ohms is for all practical purposes, close enough to 0 ohms for this discussion! 🙂
So there you go…
We’ve made the output of our vacuum tube stage beefy enough that we can connect it to almost anything down the signal path… another Opamp stage, or a make-up Gain stage, or even an output transformer!
Looks like we’ll have a Tube preamp with Output Transformers! That’s enough mojo to take the whole gang to flavor town! Any interest in a DIY Kit? Let me know in the Facebook comments section.
On the next blog issue, we’ll talk about harmonics and compare our vacuum tube stage to a solid state/non-vacuum tube stage. Is there really a big difference? We’ll find out!
That’s all for now. I hope you like this blog article, and be sure to LIKE and FOLLOW our Facebook page so you’ll get notified of future articles.