Aleph simulation- strange waveforms

[SteveG]

So, I've been simulating the Aleph30 (with a few adjustments) to get somewhat familiar with the current source setup. I want to double the output stage and feedback and use this for an AlephX.
So, I turn up the frequency on the input to 20Khz, and the waveforms start to get a wierd kind of slant to them. Can anyone help me out as to the cause of this, and how I can improve my design? I've attached the circuit and the waveforms. THanks.
Steve

[SteveG]

here's the circuit:

[jleaman]

Hey steve what program is that .. id like to try it

Jason canyou post a link

[AudioFreak]

Well that slant looks like the original waveform with even harmonics, mostly 2nd harmonics.

[janneman]

Yes, it seems from the waveform that the gain in pos or neg going direction is different. Can you re-sim with say a triangle, and decrease the feedback to increase gain and exaggerate the effect? Maybe we can see it a bit better then.

Cheers, Jan Didden

[tortello]

What about the bias current?
Respect with the original Aleph design, it seems that the resistor that helps to build the "fixed part" of the variable current source ( you can find it between the Q2 base and the R6-R7 node) is missed.
Anyway this resistor is not needed, if you set the quiescent bias current only by the R8 voltage drop, this is not the problem...

[tortello]

I've just seen that the B+ and B- rails are about 16V: in this case the output voltage of about 14V is too high... at least it seems to me

[wuffwaff]

Steve,

there's some mistake around R9 and R10. One of them must be left out! Connect the output to R9 and throw R10 out.

william

[phase_accurate]

Hi William

While the differential input stage would indeed work with just one of these resistors in place, it is definitely never false to have the same load on both branches of a differential amplifier. I do admit that the load seen by the MOSFET - where the signal is taken off from - does not only see the resistor as load but also the input impedance of the mosfets it drives. But I don't think that the presence or absence of the second resistor would cause the anomalities seen above.

But this is just a small detail compared to your suggestion to connect the gates of the output MOSFETs to the other input MOSFET's drain => THIS WOULD DEFINITELY GENERATE A NICE OSCILLATOR!


SteveG

If the amp's bias is O.K. then the effect might be caused by slew-rate limitation. Being an asymmetrical circuit it might of course have asymmetrical sle-rate capabilities causing the asymmetry seen. If you run your simulation with a rectangular input signal you might probably see different slew rates for the rising and falling edges.

Remember that a given output signal of sinusoidal shape will call for a slwe rate of more than

2*PI*frequency[Hz]*vpeak[V]

The result comes in Volts/second. You have to divide it by one million to get the usual Volts/microsecond.

Regards

Charles

[SteveG]

JasonL,
It is Circuitmaker student edition- freeware, but limited to 50 components. Seems easy to use. I can't remember the address- try searching google.


I will try to exaggerate the effect when I get home tonight(I don't have the circuit on this computer). In it's worst form, it was the waveform plus a smaller double hump on the trailing edges. The only thing I can think of is that there is something asymmetrical as far as impedances go with the input and the feedback arrangement. There is no attempt to make the input impedances equal in this circuit, as it will be completely symmetrical when it is finished (alephX). That may be the whole problem. I hope that this is worked out when I add the other half of the output stage, and use the X feedback arrangement. It would be nice to know that it works before I go and build it, if you know what I mean. I know that it is swinging close to the rails right now. Maybe that is some of the problem. At least with this attempt I have been able to test the Aleph current source and adjust the AC current through the feedback resistors, etc. as in Zen Variations Part 2. That was my goal anyway. I guess I just started to worry when I didn't get a nice looking waveform. Anyway, I would like to know what is causing the problem, in case I end up building an Aleph someday. Also, just for the sake of learning something. I appreciate all the help... please keep the suggestions coming.
steve