Electronics Workbench 4 opamp slew rate bug.

[Graham Pratt]

A friend gave me his old EB4 a few months ago because he never used
it, and I finally got around to playing with it the past weekend.
(Full-blown SPICE etc is too hard for me!) Anyway, I had a nice little
opamp circuit chugging along driving a source follower, and the zero
voltage point of the opamp supply rails were referenced to the source
follower output. This method is interesting because the demands on the
opamp are very low as regards slew rate etc. Anyway, I found out that
this simulator measures the slew rate of the opamp with respect to
earth, not with respect to it's supply rails which in this case are
going up and down almost the same amount as the opamp's output.

Do other simulators have this problem?

 

[Kevin Aylward]

A friend gave me his old EB4 a few months ago because he never used
it, and I finally got around to playing with it the past weekend.
(Full-blown SPICE etc is too hard for me!) Anyway, I had a nice little
opamp circuit chugging along driving a source follower, and the zero
voltage point of the opamp supply rails were referenced to the source
follower output. This method is interesting because the demands on the
opamp are very low as regards slew rate etc. Anyway, I found out that
this simulator measures the slew rate of the opamp with respect to
earth,

Simulators don't "measure slew rate" in any particular manner at all. A
simulator simply solves equations.

ot with respect to it's supply rails which in this case are
going up and down almost the same amount as the opamp's output.

Do other simulators have this problem?

I'm really having trouble following exactly what your compliant is, but
I'm reasonable confident your issue has nothing to do with the simulator
itself, i.e. there is no "bug" and there is no "simulator problem".

Simulators use models. By and large, the models have little to do with
the simulator, other than they run in it. If you use a bad model, you
will get bad results. My best guess at what you are saying here, is that
you are getting incorrect results because the op-amp power supplies are
moving about. If so, this is probably because the *model* not the
simulator, is not modelling the full effect of the power supplies, e.g.,
PSRR. For transient PSRR runs, I doubt if any op-amp model does this.
Its just not something that gets done that often.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

 

[Graham Pratt]

you are getting incorrect results because the op-amp power supplies are
moving about. If so, this is probably because the *model* not the
simulator, is not modelling the full effect of the power supplies, e.g.,
PSRR. For transient PSRR runs, I doubt if any op-amp model does this.
Its just not something that gets done that often.

Hi Kevin. Imagine it this way. Suppose we have an opamp that will slew 1v/uS.
If both of it's power supply rails are equally moving downward at -0.8v/uS
wrt ground then we would reasonably expect the opamp output to be able to
slew downward at -1v/uS wrt it's supply rails, that is -1.8v/uS wrt ground.
However the model limits the simulation to -1v/uS wrt *ground*.
Hope that clarifies things.

 

[Kevin Aylward]

Hi Kevin. Imagine it this way. Suppose we have an opamp that will
slew 1v/uS. If both of it's power supply rails are equally moving
downward at -0.8v/uS wrt ground then we would reasonably expect the
opamp output to be able to slew downward at -1v/uS wrt it's supply
rails, that is -1.8v/uS wrt ground. However the model limits the
simulation to -1v/uS wrt *ground*. Hope that clarifies things.

Sure, as I implied, if the model ties various bits and pieces to a
ground that isn't really there, then the model is going to fail on some
set-ups. The point I am making is that this is nothing to do with the
simulator. Its not a simulator bug. Its an inaccurate model. Its not
"...the simulator measures the slew rate of the opamp with respect to
earth...". This statement makes no logical sense. a simulator does what
it is told to do.

So, there is no "simulator problem", which you asked the question about.
The issue is in obtaining an accurate .subckt model that models slew
rate correctly under all conditions.

Kevin Aylward
[email protected]
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.