Am 09.07.2023 um 05:52 schrieb Bell, Dave <Dave.Bell@...>:

eewiz, I think Bonkers was mixing absolute timing with relative timing notation.
If you delete all of the + signs at the beginning of the lines, it makes pretty good sense, as?absolute times.
?
It took me a long time to get it through my head that relative time notation made the job a lot?easier!

Dave
?
From:?[email protected]?<[email protected]>?On Behalf Of?eewiz
Sent:?Saturday, July 08, 2023 6:34 PM
To:[email protected]
Subject:?EXTERNAL: Re: [LTspice] Discrete data points in plot view
?
Hello Christoph,
?
Bonkers may be leading you down the primrose path.
If you follow Bonkers' advice the length of bit[1] will be twice the length of bit[0] and the length of?bit[2] will be three times the length of bit[0].
The slew rate will also progress to 2 times and then 3 times longer...
?
This is all you need to make a PWL file with relative timing.
0 ? ?0 ? ? ? ? ? ? ? at 0 time its 0 volts.
+750n ? ?10 ? ? take 750ns to go from 0V to 10V with a linear slope.
+100u ? ?10 ? ? stay at 10V for 100us.
+1.5u ? ?-10 ? ? take 1.5us to go from 10V to -10V with a linear slope.
+58u ? ?-10 ? ? ?stay at -10V for 58us.
+1.5u ? ?10 ? ? ?take 1.5us to go from -10V to 10V with a linear slope.
+100u ? ?10 ? ? stay at 10V for 100us.
Rinse and Repeat...
?
If you would prefer to work with parameters instead of numbers, be my guest.
It took me just several minutes to copy and past what I needed in Excel to create a file that's?14,582 lines long.
As I made it, I made some changes to add some discontinuities on a periodic basis and then?paste, paste, paste some more.
My file runs for about 583ms.
If I had known of the REPEAT syntax at the time, I could have made use of it but, cut and paste is?easy, LTspice syntax is not.
?
Bonkers wrote:
Hi Christoph,?
Firstly, you can easily repeat cycles of your 32 points, even just using Notepad, if you?adopt the +xxx option for the timestamp (relative timing).?
Secondly, PWL means it is linearly interpolated between the defined points. ?If you?want "crisp" edges, then you need excel, or libreoffice calc, or whatever.?
Essentially you need 2 PWL values per point, to get nice square data.?
The time deltas are normally just 2 values - bit period, and slew time.?
so, to make a datastream, like "101" :?

time ? ? ? ? ? ? ? ? ? ? ?value
0 ? ? ? ? ? ? ? ? ? ? ? ? ? 0
+tslew ? ? ? ? ? ? ? ? ? bit[0] say 5V
+tbit ? ? ? ? ? ? ? ? ? ? ?same
+tbit+tslew ? ? ? ? ? ?bit[1] say 0V?
+2x tbit ? ? ? ? ? ? ? ? ?same
+2x tbit +tslew ? ? ?bit[2] say 5V
+ 3x tbit ? ? ? ? ? ? ? ? same
+ 3x tbit ?+tslew ? ? next bit
...
etcetera.?
really easy to do in a spreadsheet, then select the 2 columns, paste to notepad, save?as PWL.
All for now

...?that the cat remains both alive and dead until the state has been observed.

Tim

If I may chip in here and finish the little ditty you started with the rarely known second line John:

"Curiosity killed the cat,
but satisfaction brought it back!"

Yes, I did mean 'will'. My (limited) experience is that even different data books don't give the same graphical data, and models are no different in that respect.? Yes, you can tweak a model, but most people don't have your insight into the arcane parameters that are used in models.

I agree about diodes being calculable, as long as the ohmic resistance component is included, which is rarely specified for small-signal diodes.

Curiosity, of course, killed the cat. ?

I presume you meant "It's unlikely that a model "will* produce curves that closely match those in the data sheet"? Of course that's not true. In many cases, you can tweak a model to produce almost overlay what's in the datasheets. Why not make the model as good as you can? Don't give the "simulations can't be trusted brigade" more ammunition.

And there again, there are many cases where the model is nowhere "close enough", even for diodes. Basic discrete diodes are quite simple to optimise as there are relatively few model parameters. You don't even need LTspice for this, it can easily be done in a spreadsheet using its internal solver. There's nothing very mysterious about the diode equations.

As far as understanding what each parameter does, LTspice gives us the tools to analyse them: just change one and see what it does. Whatever happened to curiosity?

--
Regards,
Tony

On 10/07/2023 17:58, John Woodgate wrote:

No doubt you can carry out an optimization, but it requires a deep knowledge of what each of the model's parameters determines, which is far from straightforward.

It's unlikely that a model with produce curves that closely match those in the data sheet, and determining whether a match is 'close enough' also requires deep understanding unless the 'match' is a gross mismatch.

John wrote, "...?an inverse FFT will produce a (fictitious) waveform whose spectrum is the frequency response. I can't see much use for that."? It might be useful to see the impulse response of your network.? Why do you say fictitious?

And, "But an FFT of an FFT is a cepstrum, and in some fields, cepstrum analysis if found useful."? That is a term with which I was not familiar.? Mike Engelhardt recommends it, on his Help page about Exporting/Merging Waveform Data.

Anyway -- I was just trying to explain why you might have found the *.imp file in your folder.? Doing an FFT on an .AC sweep seemed more likely than accidentally doing an FFT of an FFT.

Andy

No doubt you can carry out an optimization, but it requires a deep knowledge of what each of the model's parameters determines, which is far from straightforward.

It's unlikely that a model with produce curves that closely match those in the data sheet, and determining whether a match is 'close enough' also requires deep understanding unless the 'match' is a gross mismatch.