Help with LTSpice... or something else?

[shakeshuck]

Try grounding the load on the secondary, also why do the transient and AC analysis have different nodes?

The grounding seems to have done the trick - the wave gain behaves more as expected now.

I'm afraid I'm not quite sure what you're asking in the second part of your question; I was comparing input and output wave shapes to see the distortion when I noticed the odd gain changes.

Is there an easier way to find what distortion types are in effect, e.g. harmonic, IM, etc?

I took a look at the FFT output but I don't know how to interpret it.

 

[shakeshuck]

Found the .four command.

I can see the mathematical view of harmonics now!

 

[jazbo8]

One is reading V(n006) while the other V(n007).

 

[shakeshuck]

I was comparing the harmonics tables between models this morning, and finally twigged what the FFT chart is showing me!

Great stuff. Thanks for the heads up!

 

[shakeshuck]

Using LTSpice to learn is not going well

I have a new issue, and once again I'm not sure if it's down to my misunderstanding of circuits or LTSpice.

For the circuit below (not complete!) I have displayed what LTSpice is showing me the waveforms are at various points.
I was expecting to see wav-1 and wav-1-inverted to be zero-based and show a wave of approx +/- 12v. I also expected the waves for plate-1 and plate-2 to be the same, but inverted.

Is it me not understanding the circuit, or LTSpice needing some help to find it's feet?

 

[jazbo8]

Your are missing the bias for the output tubes... as well as the grid resistor for U2.

 

[shakeshuck]

Yes, I said it wasn't complete

I wanted to see the voltages then I could take a stab at the bias - but what is shown by LTSpice is relatively useless...

 

[jazbo8]

Garbage in, garbage out...

 

[shakeshuck]

Thanks for the explanation.
Very Helpful.

 

[Mooly]

What jazbo is saying is that LT is correctly showing the result i.e. a non functional circuit because vital parameters are not set.

Try biasing just the phase splitter first (with it not connected to the 6L6's)

 

[jazbo8]

Don't blame the tool...

 

[shakeshuck]

It took me a while to twig that even though it's connected to grid, as far as LT is concerned it's still an open circuit...

 

[jazbo8]

What happens when the power tubes' grids are at 0V? Are they still "open circuit"?