Re: OPA891 / OPA2891 Model Needed
On Sat, Dec 21, 2024 at 04:56 PM, <info@...> wrote:
It is a pain to copy the model to every directory with an .asc in it and add the .lib directive to every simulation.? I really want to be able to add the symbol to the schematic and have LTspice find the proper model by itself.
I understand that and it's easily feasible, by creting your own directories. I used to do that but you run the risk of making your project not portable if you forget to join all the necessary files (models and symbols) needed to run it. I have long ago decided to put them in the same directory as the .asc, at the cost of having to copy numerous files.
Just my $0.02...
|
Re: Confused, adding a pin to an existing Symbol and .cir
You've forgotte,n to put the symbol and model in your zip file.
Le 23/12/2024 ¨¤ 18:25, Bell, Dave via
groups.io a ¨¦crit?:
toggle quoted message
Show quoted text
No longer confused (I hope!), I uploaded
a sample circuit I was working on.
The main circuit path was from an Analog
Dialogue example, useful for a bench function generator gain
controlled output.
Once I had a dynamically controlled Pot,
I also wanted to work out how to control another one,
separately.
?
Dave
?
|
Re: Ferromagnetic core modeling.
On Sun, Dec 22, 2024 at 10:19 PM, Andy I wrote:
(Note that the "A" dimension on Fair-Rite's datasheet is not the "A" parameter in the Chan model.? You probably knew this already.)
No, Andy I haven't gotten that fare yet. I am still studying Alexanders files. Thanks for the "heads-up".
|
Re: Confused, adding a pin to an existing Symbol and .cir
No longer confused (I hope!), I uploaded a sample circuit I was working on.
The main circuit path was from an Analog Dialogue example, useful for a bench function generator gain controlled output.
Once I had a dynamically controlled Pot, I also wanted to work out how to control another one, separately.
?
Dave
?
|
Re: Confused, adding a pin to an existing Symbol and .cir
As Andy suggested, I do want the pot to be dynamic.
I tried several of the many designs, and that was the most common problem; I could use a .PARAM, but not a control variable, V(), I(), or time.
?
I did eventually realize that a separate subckt, pot2, was useless. And the label vpct was
not ¡°just¡± a comment.
It¡¯s in the SpiceModel line, and does get passed to the .cir, as the value of Vpct.
Adding a second pot requires changing Vpct to Vpct2 for example, and providing a net, Vpct2.
(And RTot in SpiceLine, if required¡)
?
Dave
?
toggle quoted message
Show quoted text
From: [email protected] < [email protected]> On Behalf Of
Andy I via groups.io
Sent: Monday, December 23, 2024 5:18 AM
To: [email protected]
Subject: EXTERNAL: Re: [LTspice] Confused, adding a pin to an existing Symbol and .cir
?
On Mon, Dec 23, 2024 at 03:57 AM, Tony Casey wrote:
Still seems a lot of trouble to go to for each pot. I'm curious what's wrong with Helmut's old (but excellent) pot library? That uses a wiper position, which can either be a numeric (default 0.5), or you can replace it with a parameter,
which is unique to each pot.
Maybe Dave needed to control the pot using a voltage, so that it can change dynamically.
I think there must be a dozen or more (perhaps dozens of) potentiometer symbols and models already.? Surely some of them might already use a control voltage that is not a hidden node.? If that was the aim, even if one did not exist already,
it should have been easy to make one.
I think Dave's attempts got several things wrong.? I think he tried to use a Comment (Vpct, Vpct2) as if it were a parameter (Comments are always ignored), and he defined two subcircuits with different names (pot, pot2), when in fact they
used the same symbol so both instances actually have the same name (pot), making the second subcircuit definition meaningless as there was nothing to call it.? I think it would have been simpler to just add a 4th pin to the pot symbol for connection to the
controlling voltage node.? Why make it harder than it needs to be?
|
Re: Ferromagnetic core modeling.
On 23/12/2024 14:30, Andy I via
groups.io wrote:
vOn Mon, Dec 23, 2024 at 04:06 AM, Tony Casey wrote:
For those not listed, there is Character Map.
But again, that's Windows.
?
And anyway, I might be wrong but I thought Windows dropped
the Character Map tool, another one of its dumb ideas.? I saved
my copy so I can still use it.
you are wrong...
.. well I have "Mapa de Caracteres" on this Windows/11 laptop, but I
see its an optional add-in. I use it from time to times as laptop
does not have a numeric pad so to type directly? you have to use the
on-screen keyboard.
?
Personally, I don't mind when I see "u", which was in the
LTwiki.? LTspice lets us use "?"
but sometimes that gets us into trouble.? That happened a lot in
this group when it was on Yahoo!Groups due to its online
editor.? I think Helmut changed his LTspice settings to not use
"?",
to avoid that problem.
?
Andy
?
Dave
_
|
Re: Ferromagnetic core modeling.
I wrote:
And anyway, I might be wrong but I thought Windows dropped the Character Map tool, another one of its dumb ideas.
The "dumb idea" was the decision to drop the tool, not the tool itself.? Sorry for any misunderstanding.
?
Andy
?
|
Re: Ferromagnetic core modeling.
vOn Mon, Dec 23, 2024 at 04:06 AM, Tony Casey wrote:
For those not listed, there is Character Map.
But again, that's Windows.
?
And anyway, I might be wrong but I thought Windows dropped the Character Map tool, another one of its dumb ideas.? I saved my copy so I can still use it.
?
Personally, I don't mind when I see "u", which was in the LTwiki.? LTspice lets us use "?" but sometimes that gets us into trouble.? That happened a lot in this group when it was on Yahoo!Groups due to its online editor.? I think Helmut changed his LTspice settings to not use "?", to avoid that problem.
?
Andy
?
|
Re: Confused, adding a pin to an existing Symbol and .cir
On Mon, Dec 23, 2024 at 03:57 AM, Tony Casey wrote:
Still seems a lot of trouble to go to for each pot. I'm curious what's wrong with Helmut's old (but excellent) pot library? That uses a wiper position, which can either be a numeric (default 0.5), or you can replace it with a parameter, which is unique to each pot.
Maybe Dave needed to control the pot using a voltage, so that it can change dynamically.
?
I think there must be a dozen or more (perhaps dozens of) potentiometer symbols and models already.? Surely some of them might already use a control voltage that is not a hidden node.? If that was the aim, even if one did not exist already, it should have been easy to make one.
?
I think Dave's attempts got several things wrong.? I think he tried to use a Comment (Vpct, Vpct2) as if it were a parameter (Comments are always ignored), and he defined two subcircuits with different names (pot, pot2), when in fact they used the same symbol so both instances actually have the same name (pot), making the second subcircuit definition meaningless as there was nothing to call it.? I think it would have been simpler to just add a 4th pin to the pot symbol for connection to the controlling voltage node.? Why make it harder than it needs to be?
?
Andy
?
|
Re: Ferromagnetic core modeling.
On 23/12/2024 00:18, John Woodgate
wrote:
You
can usually write '?' in email and elsewhere using ALT-0181 on the
numeric keypad.
If you use Thunderbird as your email client, there are a number of
characters and symbols you can add from the menu item: Insert >
Characters and Symbols > Common Symbols. The drop-down list
contains "?".
For those not listed, there is Character Map.
--
Regards,
Tony
|
Re: Confused, adding a pin to an existing Symbol and .cir
On 23/12/2024 00:06, Bell, Dave via
groups.io wrote:
Never
mind, I finally figured it out:
I needed to change only the Spice Model line, to make it unique to
the instance.
Then, the hidden control voltage must be that same unique name.
Dave
Sorry ¨C I forgot to click ¡°Notify users¡± in the upload:?
Vpct_pots.zip
Attempting to modify an archive Pot component, to add a wired Pin
instead of a hidden net name for control input.
In order to use more than one pot in a circuit, the SUBCKT
definition and instance parameters need to be modified for each
pot placed.
I¡¯d like to add a pin, and change the SUBCKT definition to use
that pin rather than a hidden net name.
I¡¯ve added a pin (4), removed the pin label ¡°Vpct¡± and used V(4)
to set the value of B1, but that¡¯s not working.
When Run, LTspice complains that the pin count of the instance
doesn¡¯t match the definition.
The uploaded zip has a working model, with the separate embedded
SUBBCKT definitions.
How can I make the mod, to add a control pin and have one
includable .cir ??
Still seems a lot of trouble to go to for each pot. I'm curious
what's wrong with Helmut's old (but excellent) pot library? That
uses a wiper position, which can either be a numeric (default 0.5),
or you can replace it with a parameter, which is unique to each pot.
--
Regards,
Tony
|
Re: LMH6321 Buffer Not Working
It is now working properly.
Helmut had an LMH6321.zip demo in the files section. His model was an older one that was only slightly different, but I was able to make my model and asy files work together.
Thanks for your help, still learning.
Steve
?
toggle quoted message
Show quoted text
On 2024-12-22 18:23, Craig Lambert via groups.io wrote:
Make sure your symbol pins are numbered in consecutive order: 1, 2, 3, 4 works, 1, 3, 4, 5 (or any other sequence with a missing number) doesn't, even though the number of pins is correct.
TI supplies a seemingly straightforward spice model for its LMH6321 Unity Gain Buffer.? It requires a modified opamp symbol with only 4 nodes, (IN V+ V- OUT).
I edited a standard opamp symbol to have only the 4 nodes and get the resulting error message:? "Fatal Error: Port(pin) count mismatch between the definition of subcircuit "lmh6321" and instance: "xu1"
The instance has more connection terminals than the definition."
?
I count only 4 pins on the model and 4 pins on my symbol.? Can you see what I am doing wrong?? I will upload my files to temp.
?
Thanks,
Steve Hogan
?
?
?
?
?
?
?
?
?
|
Re: Looking for ideal fully differential amplifier spice model
On Sun, Dec 22, 2024 at 10:50 PM, Tom wrote:
?
Specifically, here:
Photos > Delta-Sigma Converter > 2024-12-22_224815.jpg
?
Andy
?
|
Re: Looking for ideal fully differential amplifier spice model
Screen shot of Delta-Sigma Converter uploaded
|
Re: Looking for ideal fully differential amplifier spice model
It's primary use is for XSPICE mixed-mode simulations. For XSPICE the key is simplicity so simulation times are reasonable. I needed something quick to replace the original code model used in a Sigma-Delta converter example. I look forward to better models with user configured parameters.?
|
Re: Whither 4053_tercio.asy
I located the missing pieces.
All of the examples work now.
Try again, please.
?
Sent:?Sunday, December 22, 2024 at 12:29 PM
From:?"Jerry Lee Marcel via groups.io" <jerryleemarcel@...>
To:[email protected]
Subject:?Re: [LTspice] Whither 4053_tercio.asy
?
Le 22/12/2024 ¨¤ 15:04, eewiz via groups.io a ¨¦crit?:
Hello John W.
I have uploaded the complete Robertugo/SwitchAna files from my personal archive.
It appears to contain all that is needed for the 4051, 52 and 53 parts.
Except the model and symbol for CD4024 that are needed to run most of the circuits!
?
|
Re: Ferromagnetic core modeling.
On Sun, Dec 22, 2024 at 05:16 PM, Richard Andrews wrote:
I am reading . "A is the bead magnetic cross section in square meters, use dimensions C*(A-B)/2, area is in mm2 hence the u".
?
?
So, is the area in meters, micrometers, or millimeters?
Neither.? Square meters, or meters squared.
?
See your LTspice Help page for Inductors:
?
A = Cross sectional area, Units = meter**2 (meters squared).
?
If even says so in the text you quoted from the LTwiki: "A is the bead magnetic cross section in square meters".
?
Fair-Rite's datasheet uses millimeter dimensions so needs to be converted to meters.? Computing C*(A-B)/2 to get area in meters^2 means adding a 'u' multiplier to the computed area when using Fair-Rite's datasheet dimensions.? (Note that the "A" dimension on Fair-Rite's datasheet is not the "A" parameter in the Chan model.? You probably knew this already.)
?
Andy
?
|
Re: Separate from .CKT for .lib file
Also -
?
What is this .CKT file you keep talking about?
?
Presumably your SPICE model for the part is in the .LIB file, and only in the .LIB file.? If that is the case, then the .CKT file is not part of the model.? Maybe the .CKT file is something needed by PSpice or some other simulator.? LTspice doesn't need that file.
?
I'm not saying that LTspice never uses a .CKT file.? If the part's SPICE model was in the .CKT file, then that is the file LTspice needs.? It doesn't care what is the filename extension.? But because you also have a .LIB file, chances are very good that the .LIB file is the SPICE model and the .CKT file is not.
?
It would not be unheard-of to have parts of a model in both files.? Very unusual, but not impossible.
?
The best way to tell is by looking at what's inside each file.
?
Andy
?
?
|
Re: Separate from .CKT for .lib file
On Sun, Dec 22, 2024 at 08:35 PM, Mirza wrote:
Can I find any example where the lib file is separated according to the components?
It would be pointless for anyone to show you an example like that.? Every circuit is unique.? "All" you need to do is identify the parts that belong to the photodiode and to the amplifier, and delete one of them.? Deleting things from the model file means literally deleting lines of text.
?
Since you can't share the model file with us, there is no way anyone here can do that for you.? The only way to tell which parts belong to the photodiode and which parts belong to the transimpedance amp, is to actually SEE it for ourselves - which we can't do.? You are on your own.? Sorry.
?
Chances are good that the photodiode part is smaller, so it might be easier to single it out.? It might be as simple as one line in the model file, or it could be dozens of related lines.
?
I meant I do not understand the meaning of contents o fckt or lib file.
For example, there are hundreds of lines and each lines have meaning.
By seeing these lines, I do not understand and also which lines belong to photodiode and which lines belong to transimpedance amplifier.
Then in that case, I think you are in trouble.? Can you find another engineer where you work who understands SPICE netlists and is covered by the NDA and can help you with this?
?
SPICE netlists are not hard to figure out.? For example, any line starting with "R" is a resistor.? "D" is a diode.? The first letter is always the key.
?
Andy
?
|
Re: LMH6321 Buffer Not Working
On Sun, Dec 22, 2024 at 05:54 PM, <info@...> wrote:
I edited a standard opamp symbol to have only the 4 nodes and get the resulting error message:? "Fatal Error: Port(pin) count mismatch between the definition of subcircuit "lmh6321" and instance: "xu1"
The instance has more connection terminals than the definition."
?
I count only 4 pins on the model and 4 pins on my symbol.? Can you see what I am doing wrong?
I think there is nothing technically "wrong" with the symbol file and model you uploaded.? I created a schematic and added your symbol to it (after editing the file's location since that is unique to your computer), and the simulation ran without errors.
?
However, you did use the SpiceModel attribute, which is NOT what I recommended using.? You should have used the ModelFile attribute instead, to specify what model file is to be loaded.? What you used was an "exception" case, and it's risky.? If you get any of the parameters wrong, it might not work the way you wanted it to work.
?
If you edited your symbol after adding it to your schematic already, you may need to delete the symbol from the schematic and then add it back (not Undo, but actually adding a new one).? There are some properties that are "sticky" and may remain even if you later alter the symbol in the symbol editor.? The schematic might retain the old ("sticky") value.
?
Andy
?
|
Bell, Dave
Tony Casey
eewiz