How to measure the output impedance of a valve preamp?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Hi everyone,
I was trying to measure the output impedance of a DIY preamp for the past couple of days without any luck. Each time, I got to the same result of around 3K3 ohm, which is unbelievably low.

I have a rather long monologue with a bunch of pictures so im attaching a PDF here with all the schematics and my solution attempts.

Was I on the right track? What have i done wrong?
If not, then how do I calculate the output impedance of a preamp?

Any help greatly appreciated.
 

Attachments

  • froumQ.pdf
    141.9 KB · Views: 242
R3 is the anode load resistor and the value is 120k therefore the output impedance of that stage is 120k.
There are a few non standard values employed in your models so this is pure theory.
Where you get 3k3 from I am not sure unless you are measuring the 3k3 resistor effectively across the 1M0 pot. Seems odd as to why you would load a reasonably high impedance, (120k) with 3k3.
 
The output impedance of the 12AX7 plate resistance, rp, of 60k Ohm, is in parallel with the 120k Ohm plate load resistor. That is 120k/3 = 40k. (120k/2 = 60k, and that is in parallel with 120k, so 120k/3).

All that other stuff including the potentiometers and 3.3k resistor ought not to be in the circuit.

A 12AX7 plate circuit can not drive that 3.3k Ohm resistor.
 
Moderator
Joined 2011
I was trying to measure the output impedance of a DIY preamp for the past couple of days without any luck. Each time, I got to the same result of around 3K3 ohm, which is unbelievably low.
The idea is to compare the open circuit output voltage, to the output voltage when a load resistor is connected, and adjust the load for half voltage. Of course, the output impedance with the 3k resistor actually connected will be around 3k, since the circuit's 40k output impedance is much higher. The Thevenin equivalent output impedance (with the 3k connected) is the 3k in parallel with the circuit's 40k output impedance, which is still about 3k.
 
Last edited:
AX tech editor
Joined 2002
Paid Member
I don't see anything in your write-up about output impedance measurement. Here's how to do it (Spice or test bench, same method, but I will describe Spice as it is much faster ;-).

Set the input source to zero or ground the input. Put an AC current source, let's call it Itest, between output and ground. Say 0.1mA. Then plot Vout/Itest. That is Zout. Ohms Law.

But maybe first clean up so that you get a well-functioning amp, without all that low impedance load stuff. That'll kill it.

Jan
 
Last edited:
Moderator
Joined 2011
I was trying to measure the output impedance of a DIY preamp for the past couple
of days without any luck. Each time, I got to the same result of around 3K3 ohm,
which is unbelievably low.

Here's the idea.
 

Attachments

  • Rout (1).png
    Rout (1).png
    7.7 KB · Views: 365
Last edited:
By pulling the signal with 3K3 resistor to the ground I am trying to lower the output level of the preamp. I found this value just by trial and error and it brings the output to the acceptable level. Without that the preamp is way too loud. So I was wondering what output impedance I am getting after all? Please see attached the re-drawn version of the last 2 gain stages. When I am calculating the output impedance of the last gain stage alone, I am getting 40K which seems logical. But when I am trying to calculate the total impedance at the output of the circuit, I am getting around 3.25K which doesn't make any sense to me. How come Z out became 10 times smaller after all the passive components that create an extra load on the output signal?:confused:
 

Attachments

  • tube preamp.png
    tube preamp.png
    21 KB · Views: 361
Moderator
Joined 2011
By pulling the signal with 3K3 resistor to the ground I am trying to lower the output level of the preamp.

When I am calculating the output impedance of the last gain stage alone, I am getting 40K which seems logical. But when I am trying to calculate the total impedance at the output of the circuit, I am getting around 3.25K

You need a volume control (or fixed attenuator) at the input to lower the level. A heavy load resistor
connected in shunt on the output will cause high distortion and poor bass response.

See posts #4 and #6 for the second answer.
 
Last edited:
Right, but this is a guitar preamp, so I actually want to overdrive the gain stages. So I can't lower the input signal. I just need to quiet it down at the output. Perhaps, I am not getting it right, but to my understanding the output impedance is a sort of resistance in series to the output signal. As it is shown in Merlin's book. See the pic attached. So, based on my calculations, my Z out is getting as low as 3k3 and the output signal is meeting a very low resistance. While the logic says it is actually the opposite. The signal is heavily loaded. Just can't get my head around it.:spin:
 

Attachments

  • impedance.png
    impedance.png
    47 KB · Views: 340
The 270pf from plate to cathode makes a great low pass filter. Xc = 29k Ohms at 20kHz, and the 12AX7 plate rp and 120k load is 40k. response is -3 dB at less than 20kHz. Is that what you want? What is the intended purpose and function of your preamp? What is the signal source at the input? A microphone, phono cartridge, CD player, Tuner, etc.? What is the output of the preamp going to drive? A power amp or what?
 
Last edited:
my Z out is getting as low as 3k3 and the output signal is meeting a very low resistance. While the logic says it is actually the opposite.

The circuit's output impedance is 40k, and you are loading that with 3k to ground. The voltage divider formed by these two resistances reduces the output level a lot, more than a factor of ten.

However, the output impedance of the circuit with the 3k in place is changed from 40k to 3k. You would have to study Thevenin's theorem to have an intuitive understanding of it. Failing this, just accept that the output impedance of the circuit with the 3k load is the 40k in parallel with the 3k.

The Thevenin Theorem - A Simple Explanation - Build Electronic Circuits
 

PRR

Member
Joined 2003
Paid Member
If you truly need 3K on the output to get your signal level, then of course Zout will be <3K.

With your backward-connected Master Volume pot in the latest plan, final Zout will vary from 250k to <3K. To my mind, _I_ would think to look at a normal-connection 5K pot (possibly with 10K across to make your 3K) which would give <5K at all settings and be much less susceptible to various loadings.
 
Thank you, PRR.
So, I got the concept of the output impedance all the way wrong.
Could you tell, on the most basic level, the definition of the output impedance in such circuits and how it will interact with the different loads?
I am just trying to get the basic simplified idea of it.
 
You can model the output of almost any circuit as a voltage source with a resistor in series with it. The voltage source will give whatever voltage it gives, however much you load it. The series resistor will drop some voltage if you take some output current. You can use Ohm's Law to calculate how much voltage drop. The series resistor in this model is the output impedance.
 
The post #6 shows one of the ways to calculate the output Z. Which I can totally understand. I care about the output impedance topic just because I would like to understand the concept of the correct signal matching. To my understanding the lower the output impedance is - the better. The ideal z out is 0. But I am confused by the fact that in real circuit Zout is going down when the output signal is pulled to ground. Which is not a good thing for a signal, right? I guess I am missing something between the lines.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.