Apologies for what may be a remedial question. For an amplifier simulation in LTSpice, I have a voltage source creating a sine wave for the input. In most of the other simulations I've seen as examples, this input voltage source is set to a very small Amplitude--0.3V for example, I've also seen things like 0.2V or 0.5V. However, according to the documentation for my DAC, it outputs a fixed Vrms of 4 volts. So, should I set the input sine wave to an amplitude of 4 to imitate the signal coming from my DAC? Or am I misunderstanding something fundamental (likely). I get radically different results from setting the input to 0.3 vs. 4.
In LTspice the amplitude you set is the peak voltage. So if you set the voltage as 0.3 you get a sine that goes from 0V up to 0.3v and then back down through zero and to -0.3. So the amplitude is 0.6 volts peak to peak or 0.212 volts rms. Vrms is peak voltage divided by root 2.
4 volts rms would need LTspice set to give a voltage of 4 * root 2 or 5.656 volts.
4 volts rms would need LTspice set to give a voltage of 4 * root 2 or 5.656 volts.
As an arbitrary example, here's a picture of a simulation of the SIT-5 amplifier: https://www.diyaudio.com/community/threads/first-watt-sit5.418023/page-10#post-7885828
The creator has set the sine wave to 0.45 amplitude. Why would that be the amplitude of a signal that's coming not just from a DAC, but given that this is a power amp, from a preamp that would further amplify the DAC's signal?
(Perhaps I should @ the creator @generg if they're around and don't mind)
The creator has set the sine wave to 0.45 amplitude. Why would that be the amplitude of a signal that's coming not just from a DAC, but given that this is a power amp, from a preamp that would further amplify the DAC's signal?
(Perhaps I should @ the creator @generg if they're around and don't mind)
The lower numbers all come from the signal needed to output what the person doing the sim wants.
For example... look at this circuit. The gain is set by the two resistors marked. Gain = (10000/390) + 1 which is 26.6 numerically. Those two resistors are the feedback network and they alone determine the gain.
I want an output of 2.83 volts rms which is the equivalent of 1 watt into 8 ohms. 2.83 volts rms is 4 volts peak.
Knowing the gain is 26.6 and that we want 4 volts peak (remember LT works in peak voltages, not rms) we set the input voltage to be 4 volts (our wanted output) divided by 26.6. That gives us 4/26.6 = 0.150 volts. That should be the input voltage required.
Lets see if it works.
Circuit:
We have calculated and set our input voltage to get 4 volts peak output.
And it works 🙂
For example... look at this circuit. The gain is set by the two resistors marked. Gain = (10000/390) + 1 which is 26.6 numerically. Those two resistors are the feedback network and they alone determine the gain.
I want an output of 2.83 volts rms which is the equivalent of 1 watt into 8 ohms. 2.83 volts rms is 4 volts peak.
Knowing the gain is 26.6 and that we want 4 volts peak (remember LT works in peak voltages, not rms) we set the input voltage to be 4 volts (our wanted output) divided by 26.6. That gives us 4/26.6 = 0.150 volts. That should be the input voltage required.
Lets see if it works.
Circuit:
We have calculated and set our input voltage to get 4 volts peak output.
And it works 🙂
@Mooly Thanks for the explanation! This makes sense in terms of having a target output voltage and so putting in a corresponding input voltage. From my perspective, though, what I want to do is see if I built the amplifier and plugged it into my DAC that it would produce a good result.
Wait, I just realized something that should've been obvious and is probably the source of my confusion. If I stick a potentiometer after the input, then the full 4 volts RMS isn't going to reach the gain stage because I'm not going to turn it up all the way. What I'd actually be simulating with the full 4 volts RMS is the amplifier cranked up to max, which explains why my simulations look ridiculous--the lower input values are more like the actual listening levels, and playing with them actually lets me play with the range of possible volume settings. D'oh! Alright, thanks for helping shake the cobwebs from my brain.
Wait, I just realized something that should've been obvious and is probably the source of my confusion. If I stick a potentiometer after the input, then the full 4 volts RMS isn't going to reach the gain stage because I'm not going to turn it up all the way. What I'd actually be simulating with the full 4 volts RMS is the amplifier cranked up to max, which explains why my simulations look ridiculous--the lower input values are more like the actual listening levels, and playing with them actually lets me play with the range of possible volume settings. D'oh! Alright, thanks for helping shake the cobwebs from my brain.
OK 🙂 I think you've worked it out. 4 volts rms is loads for a source and most amps will deliver full output on a fraction of that... so yes, you need a pot after the output of the DAC. In practice (real world) use something like a 10k or 20k pot rather than 100k or 220k.