Hoi Dana,

On Sun, 22 Oct 2017 09:23:26 -0500
Dana Whitlow k8yumdoober@gmail.com wrote:

As Bruce already mentioned, the simulation times required for
getting decent results is 100-1000 times as long as the lowest
frequency considered. This is basically a statistical issue as
noise simulation, to be accurate, has to average over several
"runs" to remove effects of the noise source behaviour.

Another thing is, that, for the transient simulation to be accurate,
the maximum steps size has to be limited such that the maximum voltage
or current step seen is small. Ie if there is anything in the system
with high slew rates, then the step size has to be adapted to this
slew rate. This in turn makes it slower

Additionally, most spice implementations have quite short running
random number generators (usually with a state space of 2^16 to 2^32,
few with 2^64) which in turn requires some tricks to get decent results
out of it, that again make the simulation time longer.

			Attila Kinali

--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson

Hi Rafael

On Sun, 22 Oct 2017 17:20:52 +0200
Rafael Gajanec rgajanec@vercet.com wrote:

Simplified, they are differential amplifiers driven into saturation.
A bit more detailed, I am looking at ring oscillator stages and sine-to-square
conversion circuits and their behaviour regarding various key factors
(note: I am not sure what the key factors are, yet)

Hmm.. I didn't know about Harmonic Balance. I have some reading up to do.
Thanks!

I am mostly using ngpsice, because it's very easy to script (I have a bunch
of perl scripts that feed simulations into a Grid Engine cluster, extract
data and analyzse it). Is there any big advantage of the commercial spice
engines that would make them worth considering? And would the license alow
to run hundreds of instances in parallel?
(Yes, I am doing crazy things :-)

		Attila Kinali

--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson

Hoi Bruce,

On Mon, 23 Oct 2017 09:58:40 +1300 (NZDT)
Bruce Griffiths bruce.griffiths@xtra.co.nz wrote:

Unfortunately, my understanding of transistors is far from being good.
Hence doing accurate analytical analysis is beyond me. Hence my reliance
on spice to do the "hard work."

This is a good idea, thanks!

I tried to read the yahoo groups... but god! is the interface bad!
I only found a few mails from last August that go into that direction.
Is there anything else that might be interesting?

		Attila Kinali

--
It is upon moral qualities that a society is ultimately founded. All
the prosperity and technological sophistication in the world is of no
use without that foundation.
-- Miss Matheson, The Diamond Age, Neil Stephenson

Hi Attila,

On 27-Oct-17 8:25 PM, Attila Kinali wrote:

Oscillator design - that's what I found HB simulation particularly
useful for. It gives you almost instant results, compared to the
transient simulation, say 10 seconds instead of 5 hours! Just imagine
what it means if you are trying to tune several parameters of an
oscillator... The only other reasonably fast and accurate way I can
think of is to build the bloody circuit and measure it using some
expensive equipment.

Attached are some results of a simple transient simulation using Hspice
M 2017.03, BBspice A/D 5.2.3 and ADS 2016.01. It's basically *V1 1 0 SIN
0 1 1Meg *and then .FOUR 1Meg V(1) in Hspice, VspecTran in ADS and
spectra computed using postprocessor in BBspice and ADS. As you can see,
there are some differences... To be fair, possibly there are some
simulator-specific settings/methods that could improve the results and
you should figure it out yourself what's the way to get the best results
from your spice. See
http://www.audio-perfection.com/spice-ltspice/distortion-measurements-with-ltspice.html

Commercial spice engines may have lower computational noise and shorter
simulation times. For example my out-dated BBspice (which is commercial
too by the way) crashed several times before I got some results, while
it used little RAM and only about 10-12% of available processor
resources... I intended to get you Pspice results of this simulation as
well, but I gave up after half an hour and about 1% of progress.

Best regards,
Rafael Gajanec

Hi

The “fun part” of harmonic balance is making sure you are not off in a corner
case where the results are not as good as they might otherwise be. Maybe not
as much an issue for a VCO as for some other structures.

Bob