Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Damping factor of amplifier in LTSpice simulation

Status
Not open for further replies.

northumber82

Member level 3
Joined
Jan 6, 2018
Messages
66
Helped
2
Reputation
4
Reaction score
2
Trophy points
8
Activity points
564
Hi, I've to measure the damping factor of my amplifier using LTspice simulation.

I've actually connected a voltage source to the output in with a 800ohm series resistor and grounded the amplifier input. The DC transfer function show me this result:

Screenshot (3).png

It's correct what I've done? How to calculate now the damping factor?
 

Damping factor of which transfer function ?

You have performed the setup to find the Output impedance, but not exactly the output impedance because you connected additional resistance.
 

hi,

You have performed the setup to find the Output impedance,
With audio power amplifier the terminology of "damping factor" often is used for: nominal_load_impedance / amplifier_output_impedance.

Then the setup for finding the output impedance is the way to go.

***

We need more information from the OP.

Klaus
 
  • Like
Reactions: CataM

    CataM

    Points: 2
    Helpful Answer Positive Rating
Ok probably I have solved myself. Adding AC voltage source to the output, measuring impendance at "ou1" and "ou2" poles. This is the result:

Screenshot (4).png

Then I have a damping of about 12.7k? (800 / 0.0627)
 

Hi,

from your measurement the values look OK.

***
I wonder: Why you have a such powerful, paralleld output stage when the nominal_load is 800 Ohms?
Isn´t this overkill?


Klaus
 

I wonder: Why you have a such powerful, paralleld output stage when the nominal_load is 800 Ohms?
Isn´t this overkill?

In what sense is an overkill? Looking in "Designing power amplifier" of Bob Cordell, when simulating the power amplifier for the DF he uses a 800ohm load. What I've to use?
 

Hi,

finally we know that you talk about audio amplifiers. Good to know.

Read the book again and read my post#3 again.
While DF can be measured with the use of a (dummy) load resistor of 800 ohms -- You surely need to use the nominal_load_impedance to calculate DF.

Klaus
 

The damping factor is the load impedance divided by the amp output impedance, so that value is meaningless unless you use the load impedance the amp was designed for.
For an audio amp that's typically 2, 4, 8, or 16 ohms.
 

Then if I use a 8ohm nominal load impedance I obtain only 127 of DF. Is too low right?
 

Hi,

You referred to Bob Cordell's book and how to calculate DF.
But it seems you didn't go to his example with 800 Ohms (measurement) load ...and he gets a DF of 100.

10V to 1mV gives a factor of 10,000.
This is the DF when you calculate with 800 Ohms as load.
But he calcultates DF to be 100, because he doesn't refer to 800 Ohms load, but to 8 Ohms nominal load.
This gives another (correcting) factor of 8/800 which is 0.01.
10,000 x 0.01 = 100

Klaus
 

Then let's see if I have correctly understood.

Immagine.png

I've attached a voltage source to the end in series of a 800ohm load resistor, as the book.

I've grounded the amplifier input.

I've done a transient analysis.

Without the voltage source, the measured output before degeneration resistor (out1 node) is about 14mV.

With the voltage source I obtain this:
Immagine2.png

I've to see only the voltage swing, right? Then is about to 1mV.

With the voltage source and a nominal load of 8ohm I obtain this:
Immagine3.png

I've done it right?
 

Without the voltage source, the measured output before degeneration resistor (out1 node) is about 14mV.
It's DC offset, irrelevant for damping factor. You will be only measuring AC voltage.

I've to see only the voltage swing, right? Then is about to 1mV.
0.55 mV peak-peak, I believe. Corresponding to 0.22 ohm output impedance.

With the voltage source and a nominal load of 8ohm I obtain this.
Voltage swing is roughly factor 100 of the previous measurement (= about same output impedance), but the waveform is rather distorted. Suggests that either class AB bias current or loop gain is insufficient.

I see that Cordell performs a DF measurement with 10 Vrms and 100 ohm series resistance, makes 100 mArms test current, roughly the same magnitude used in your 1 V/8 ohm test (88 mArms). Deriving DF for an amplifier with this distortion amount makes however little sense, you should probably start with THD measurements and fix the nonlinearity.
 

Then I continue to understand less and less. I still not understand how to measure this damping factor, can anyone explain me step to step:
1) What simulation I have to perform
2) Where to put voltage source, his voltage and frequency.
3) What series resistor to use, considering is a audio power amplifier.
4) How to put measurement nodes and what to take

FvM, the distortion at 1kHz@8ohm - 50W is perfect and ultralow.
Screenshot (6).png
 

If you copy Cordell's measurement method, you'll use 10 Vrms (14.1V sine) and 100 ohm. Measurement to be repeated at different frequencies. Measure AC RMS voltage at the output node, calculate Zout = Vout/0.1A respectively DF = 8ohm*0.1A/Vout.

Or presume linear amplifier behavior, simply perform an AC sweep with 0.1A (or 1A) AC current injected into the output.

the distortion at 1kHz@8ohm - 50W is perfect and ultralow.
Fine, but doesn't fit well the nonlinear output impedance showing in post #11. May be something has changed between both simulations. However, I'm not motivated to guess much about it. Either post the complete simulation circuit or find the reason yourself.
 

Fine, but doesn't fit well the nonlinear output impedance shown in post #11. May be something has changed between both simulations. However, I'm not motivated to guess much about it. Either post the complete simulation circuit or find the reason yourself.

Read better, the non-linear wave that you see, is the wave measured before the degeneration resistor (0.1ohm) and with the voltage source applied to the output, when the input is grounded.
 

Read better, the non-linear wave that you see, is the wave measured before the degeneration resistor (0.1ohm) and with the voltage source applied to the output, when the input is grounded.
"Read better"? Why did you select this useless test point? Please repeat with useful conditions, measuring at out node.
 

I told you that you should have to read better :lol: it's not V(out) but V(out1) that I've measured. I wanted to see the voltage swing to the output of the mosfet, that is 0.56mV
 

You can measure what you want, but for DF determination, you'll measure V(out). Hopefully you get an undistorted waveform at this node.
 

I concede that I should have "read better", because you mentioned probing the source nodes in post #4 and #11. However I don't see how the measurement should be related to damping factor which is simply the quotient of nominal load impedance and output impedance.

The amplifier feedback tries to achieve "zero" impedance at output node. Respectively other circuit nodes like the said source nodes show a higher AC voltage than V(out) when you inject an output current. By working of the bias current, there's also a DC offset at these nodes.
 

Hi,

From post#3:
nominal_load_impedance / amplifier_output_impedance.
Nominal_load_impedance = 8 Ohms.

Now you need to measure the amplifier_output_impedance.
With a shorted input (ignoring offset)
Amplifier_output_impedance = V / I = amplifier_output_voltage / load_current.

And... load_current = load_voltage / measurement_resistance

Thus it is not very important which measurement_resistance you use.

You will get the same DF when you use 800 Ohms, 8 Ohms or any other value that produces meaningful load current.

Klaus
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top