Re: LTspice models do not work in Qspice
Hi Tony,
"PROR_DELAY " is a line written in the model I uploaded, if you look from the 1st line you can see it, Qspice is objecting to this syntax!
Regards,
Suded
|
Re: LTspice models do not work in Qspice
Hi Tony,
Sorry, I will now and reupload
Suded
|
Re: LTspice models do not work in Qspice
On 19/04/2025 11:29, suded emmanuel via
groups.io wrote:
Many thanks, I've just uploaded the .ZIP file to temp:
Example
Buck Boost for upload to group.zip
Please
let me know if all is good
It
contains the .asc file and the UCC27282 model as a .subckt
UCC27282 symbol is missing. According to your schematic file, it is
in your "AutoGenerated" folder. You should copy that to the
schematic folder, and edit the symbol ModelFile attribute to remove
the hard-coded path.
However I can see in your schematic that you have used "standard"
LTspice VDMOS models. As I said previously, these are not portable.
I also don't see .subckt PROP_DELAY. Did you upload the wrong file?
--
Regards,
Tony
|
Re: LTspice models do not work in Qspice
Hi Tony,
Many thanks, I've just uploaded the .ZIP file to temp:
Example Buck Boost for upload to group.zip
Please let me know if all is good
It contains the .asc file and the UCC27282 model as a .subckt
Suded
?
|
Re: LTspice models do not work in Qspice
The QSPICE group is currently locked. The Qorvo website has some QSPICE support resources which may provide an answer to your question.
On Sat, Apr 19, 2025 at 04:35 Tony Casey via <antoniustrevorum= [email protected]> wrote:
On 19/04/2025 10:20, suded emmanuel via
wrote:
I have models (.suckt) that are working perfectly in LTspice,
when I import them in Qspice they dont work
Does Qspice use a different syntax?
Most (SPICE) simulators use the same general syntax, but all or most
simulators have features that are not supported in all the others.
Some features in LTspice are proprietary and hence not available in
other simulators.
Examples of uniqueness include VDMOS models and anything that uses
an A-device, like UniversalOpamps or logic gates. There may also be
a few functions that are not portable. However, unless you disclose
which models you are referring to, it's not possible to comment
further.
If you're wanting to use QSPICE, perhaps you should ask in a QSPICE
group. Perhaps it has an LTspice to QSPICE converter, I don't know.
--
Regards,
Tony
|
Re: LTspice models do not work in Qspice
On 19/04/2025 10:46, suded emmanuel via
groups.io wrote:
Can I upload the
.subckt file?
Of course. Go to Files
> Temp and click on the New/Upload button.
It would also be helpful if you could also provide a schematic that
(successfully) uses the .subckt, and any other files that are not
provided with LTspice. Put them in a zip, and upload that.
Don't forget to send another message if/when you do that.
--
Regards,
Tony
|
Re: LTspice models do not work in Qspice
Can I upload the .subckt file?
|
Re: LTspice models do not work in Qspice
On 19/04/2025 10:24, suded emmanuel via
groups.io wrote:
For example I have the following error:
No such subcircuit: PROP_DELAY
I have "PROP_DELAY " in the 1st line of the .subckt model
That's not very helpful unless you provide an example.
--
Regards,
Tony
|
Re: LTspice models do not work in Qspice
On 19/04/2025 10:20, suded emmanuel via
groups.io wrote:
I have models (.suckt) that are working perfectly in LTspice,
when I import them in Qspice they dont work
Does Qspice use a different syntax?
Most (SPICE) simulators use the same general syntax, but all or most
simulators have features that are not supported in all the others.
Some features in LTspice are proprietary and hence not available in
other simulators.
Examples of uniqueness include VDMOS models and anything that uses
an A-device, like UniversalOpamps or logic gates. There may also be
a few functions that are not portable. However, unless you disclose
which models you are referring to, it's not possible to comment
further.
If you're wanting to use QSPICE, perhaps you should ask in a QSPICE
group. Perhaps it has an LTspice to QSPICE converter, I don't know.
--
Regards,
Tony
|
Re: LTspice models do not work in Qspice
For example I have the following error:
No such subcircuit: PROP_DELAY
I have "PROP_DELAY " in the 1st line of the .subckt model
|
Re: LTspice models do not work in Qspice
|
LTspice models do not work in Qspice
Hi,
I have models (.suckt) that are working perfectly in LTspice, when I import them in Qspice they dont work
Does Qspice use a different syntax?
Regards,
Suded
|
Re: Does anyone have the model file for FinFET that will be compatible with LTspice ?
On Fri, Apr 18, 2025 at 02:32 AM, Say Die. wrote:
I Am trying for 20nm FinFET design and i have 45 and 180nm done but i cannot find any file for 20nm with 54 or any other level for ltspice.
One thing to keep in mind about transistor models, is that they are meant to describe the fabrication process.? They are inherently linked to only the FAB that you use.? A model for one FAB line does not work for another FAB line or from another manufacturer.? Normally, one gets the SPICE model from the manufacturer - not by finding something on the Internet!
?
When talking about (say) a 20nm process, it might be that everyone gets their chips from the same manufacturer made to the same specifications, and then there would be only one model that everyone uses for that step.? But don't count on it.? In general, you should ask the people who will make your chips, for their SPICE models.? Specify that you need LEVEL=X models (pick one from the list), or only LTspice-capable models, but they are less likely to understand LTspice than LEVEL numbers.? I would not count on models from anyone else to work for your 20nm chips.
?
Andy
?
|
Re: Does anyone have the model file for FinFET that will be compatible with LTspice ?
I looked online and it appears that there are freely available SPICE simulators that support Verilog-AMS (and thus Verilog-A).
Since you would have to decide which (if any) ?would suit your needs, I'd suggest doing an internet search using the string "spice simulators that support verilog-a" or something such as that.
The FinFET is typically modeled using the BSIM-CMG model which is not available in LTspice.? I do not know of a free simulator that supports the BSIM-CMG model unless it is through a Verilog-A representation.
?
BB
|
Re: Does anyone have the model file for FinFET that will be compatible with LTspice ?
The FinFET is typically modeled using the BSIM-CMG model which is not available in LTspice.? I do not know of a free simulator that supports the BSIM-CMG model unless it is through a Verilog-A representation.
?
BB
|
Re: Does anyone have the model file for FinFET that will be compatible with LTspice ?
On Fri, Apr 18, 2025 at 02:32 AM, Say Die. wrote:
also there's something called BSIM model file if you have this then it's also fine
A number of the MOSFET LEVELs are BSIM.? BSIM is not one single model, there are several, a few major levels? and several sub-levels.??Level=4 is BSIM,? Level=5 is BSIM2.? Level=7 is BSIM3.? Level=8 is BSIM3, version 3.3.0.? Level=9 is BSIM3 version 3.2.? Level=14 is BSIM4 version 4.6.1.? Level=49 is BSIM3.? Level=54 is BSIM4.
?
Some of this you could have ascertained by reading LTspice's Help.? Did you try looking there?
?
but I don't know how we use this file in LTspice.
And what "file" is that?? Do you have a model file that can be identified by its LEVEL number?? Use that number to see if LTspice can use it.? As far as I know, these MOSFET LEVELs are accepted:
?
1,?2, 3, 4, 5, 6,?7, 8,?9,?12,?14,?49,?54, 73
?
All other LEVELs are not accepted and cannot be used in LTspice.
?
But be aware that one or two SPICE programs might use different LEVEL numbers for the same model.? Sigh....
?
Andy
?
|
Does anyone have the model file for FinFET that will be compatible with LTspice ?
I Am trying for 20nm FinFET design and i have 45 and 180nm done but i cannot find any file for 20nm with 54 or any other level for ltspice.? if anyone have this file please provide it to me also there's something called BSIM model file if you have this then it's also fine but I don't know how we use this file in LTspice. Please, Reply ASAP
?
|
Re: Have anyone built black hole model already ?
On Thu, Apr 17, 2025 at 09:11 PM, <ericsson.sunshine@...> wrote:
Change the 'rise' time from long time to 1 second (the shortened sweep time), will solve all problems, no more confusion, simulation pain.
That does not solve *all* problems.? You will always have the fact that the input signal to the notch filter is a frequency sweep, so it is not one unique frequency.? Attempting to pass that through a high-Q filter will have artifacts.? They can be diminished (not eliminated) by slowing down the frequency sweep, as Tony said already.? This is very well known in filter theory.
?
And I guess 'Heisenberg's Uncertainty Principle' also apply , in the measuring of distance variation due to 'gravitational wave' of black hole. But this relation , don't know how to well describe in SPICE. By random() ???
I cannot speak to the physics or space-time aspect.? But in college an instructor told us that Heisenberg's uncertainty principle was first described, or applied, to problems involving frequency and time, e.g., Fourier spectra.? It was later (or around the same time) applied to particle physics.
?
I rather doubt that randomness has anything to do with this, SPICE or not.? The stuff we are talking about here is very deterministic (even if the measurements are imprecise) and randomness need not - should not - be applied - except as a way to blind the people you are talking to, and make them think that it is too mysterious for them to comprehend.? Don't go there.
And, I think the frequency sweep should be done well by both, PWL & PULSE, I was used to use PULSE.
Well, yes.? I was just suggesting that a PWL source is better suited for generating a straight-forward voltage ramp, than a PULSE source is.? The PULSE source has too many parameters that interact.? (Did you notice that LTspice rejected your value for its Period?)? Whereas a PWL is ideally suited for generating a voltage ramp when that is all that you need.
?
I am still missing the connection, if any, to a black hole.? This is just an audio filter, is it not?
?
Andy
?
|
Re: Have anyone built black hole model already ?
Hi, :
?
Thank you all who replied, it's nothing special , just to make some relevant between SPICE & physics, which should be able to describe by math equations.
?
The lib file can be found in folders,?
?
Change the 'rise' time from long time to 1 second (the shortened sweep time), will solve all problems, no more confusion, simulation pain.
?
And I guess 'Heisenberg's Uncertainty Principle' also apply , in the measuring of distance variation due to 'gravitational wave' of black hole. But this relation , don't know how to well describe in SPICE. By random() ???
?
And, I think the frequency sweep should be done well by both, PWL & PULSE, I was used to use PULSE.
I have no idea what's supposed in this 'Black hole' simulation, maybe just for fun, I slowly felt nothing can do day by day.
?
But , I guess will still try harder to make more simulation.
?
Thank you all again.
Best regards.
|
Re: creating gm/Id vs Id/W plot in ltspice
Change the whole file. I gave you a complete .plt file that
provides the plot you wanted (if I understood your request
correctly; there seems to be some potential language confusion).
Donald.
On 4/17/25 10:35, john23 via groups.io
wrote:
toggle quoted message
Show quoted text
[DC transfer characteristic]
{
? ?Npanes: 2
? ?{
? ? ? traces: 1 {34668548,0,"Is(M1)"}
? ? ? X: (' ',1,0.01,0.2,1.8)
? ? ? Y[0]: ('?',1,0,8e-07,8e-06)
? ? ? Y[1]: ('_',0,1e+308,0,-1e+308)
? ? ? Amps: ('?',0,0,1,0,8e-07,8e-06)
? ? ? Log: 0 0 0
? ?},
? ?{
? ? ? traces: 1 {524290,0,"-D(Is(M1))/D(V(Vgs))"}
? ? ? X: (' ',1,0.01,0.2,1.8)
? ? ? Y[0]: ('?',0,0,1e-06,1.3e-05)
? ? ? Y[1]: ('_',0,1e+308,0,-1e+308)
? ? ? Units: "ohm-1" ('?',0,0,0,0,1e-06,1.3e-05)
? ? ? Log: 0 0 0
? ?}
}
|
Tony Casey
BB