|
Hi,
?
I want to make my own block for a bidirectional mosfet. I can set up the .asc file with two mosfets in it and draw the .asy file.
?
I want to do something like add MOS=IPB107N20N3 in the PARAMS section on the Navigate/Edit Schematic Block but it moans at me about not being able to resolve the parameter.
?
.asc and .asy file, hopefully, in the temp directory in a bit as bimos.zip
?
Thanks for any help.
?
Per
|
.PARAM can only be used to pass numbers. From
the Help: To invoke parameter
substitution and expression evaluation, enclose the expression
in curly braces. The enclosed expression will be replaced with
the floating-point value.
Your .ASC is incomplete, anyway. There are
no supply voltages and no simulation directive. Your.ASY has no
attributes. Maybe that is OK, but maybe not.
On 2025-05-05 18:03, Per wrote:
Hi,
?
I want to make my own block for a bidirectional mosfet. I can
set up the .asc file with two mosfets in it and draw the .asy
file.
?
I want to do something like add MOS=IPB107N20N3 in the PARAMS
section on the Navigate/Edit Schematic Block but it moans at me
about not being able to resolve the parameter.
?
.asc and .asy file, hopefully, in the temp directory in a bit
as bimos.zip
?
Thanks for any help.
?
Per
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
Thanks for the reply. That's a big not going to happen then. :-(
?
As an aside the .asc file is 'linked' to the .asy file. It does not need a simulation directive. You drop the .asy on the schematic and off you, don't, go. Yes you could have internal supplies but otherwise you would link them in hopefully with net labels or pins/ports.
?
Either I am wrong or I think you know that one.
|
I've never heard of that procedure. Let's see
what others say.
On 2025-05-05 20:22, Per wrote:
Thanks for the reply. That's a big not going to happen then.
:-(
?
As an aside the .asc file is 'linked' to the .asy file. It
does not need a simulation directive. You drop the .asy on the
schematic and off you, don't, go. Yes you could have internal
supplies but otherwise you would link them in hopefully with net
labels or pins/ports.
?
Either I am wrong or I think you know that one.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
Not quite. LTspice also knows string parameters, which were introduced extactly for the purpose discussed here.
?
Best Regards,
Mathias
?
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only be used to pass numbers. From the Help: To invoke parameter substitution and expression evaluation, enclose the expression in curly braces. The enclosed expression will be replaced with the floating-point value.
|
NOW It does, but not all are running the latest version(s)
?
toggle quoted message
Show quoted text
From: [email protected] <[email protected]>
On Behalf Of Mathias Born via groups.io
Sent: Monday, May 05, 2025 12:38 PM
To: [email protected]
Subject: EXTERNAL: Re: [LTspice] Passing a Mosfet Value as a parameter to a hierarchical subcircuit
?
Not quite. LTspice also knows string parameters, which were introduced extactly for the purpose discussed here.
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only be used to pass numbers. From the Help:
To invoke parameter substitution and expression evaluation, enclose the expression in curly braces. The enclosed expression will be replaced with the floating-point value.
|
So the Help needs to be updated, ASAP.
On 2025-05-05 20:37, Mathias Born via
groups.io wrote:
Not quite. LTspice also knows string parameters, which were
introduced extactly for the purpose discussed here.
?
Best Regards,
Mathias
?
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only be used
to pass numbers. From the Help: To invoke parameter substitution and
expression evaluation, enclose the expression in curly
braces. The enclosed expression will be replaced with
the floating-point value.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
It's already in there. In great detail.
?
On Mon, May 5, 2025 at 10:36 PM, John Woodgate wrote:
So the Help needs to be updated, ASAP.
On 2025-05-05 20:37, Mathias Born via groups.io wrote:
Not quite. LTspice also knows string parameters, which were introduced extactly for the purpose discussed here.
?
Best Regards,
Mathias
?
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only be used to pass numbers. From the Help: To invoke parameter substitution and expression evaluation, enclose the expression in curly braces. The enclosed expression will be replaced with the floating-point value.
|
OK, so it's MY Help that needs to be updated.
I will see.
On 2025-05-05 22:02, Mathias Born via
groups.io wrote:
It's already in there. In great detail.
?
On Mon, May 5, 2025 at 10:36 PM, John Woodgate wrote:
So the Help needs to be
updated, ASAP.
On 2025-05-05 20:37, Mathias Born
via groups.io wrote:
Not quite. LTspice also knows string parameters, which
were introduced extactly for the purpose discussed here.
?
Best Regards,
Mathias
?
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only be
used to pass numbers. From the Help: To invoke parameter substitution
and expression evaluation, enclose the expression
in curly braces. The enclosed expression will be
replaced with the floating-point value.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
Well, I now have 24.1.8, but the Help still
says: The enclosed expression
will be replaced with the floating-point value.
On 2025-05-05 22:24, John Woodgate
wrote:
OK, so it's MY Help that needs to be
updated. I will see.
On 2025-05-05 22:02, Mathias Born via
groups.io wrote:
It's already in there. In great detail.
?
On Mon, May 5, 2025 at 10:36 PM, John Woodgate wrote:
So the Help needs to be
updated, ASAP.
On 2025-05-05 20:37, Mathias
Born via groups.io wrote:
Not quite. LTspice also knows string parameters,
which were introduced extactly for the purpose discussed
here.
?
Best Regards,
Mathias
?
On Mon, May 5, 2025 at 07:16 PM, John Woodgate wrote:
.PARAM can only
be used to pass numbers. From the Help: To invoke parameter
substitution and expression evaluation, enclose
the expression in curly braces. The enclosed
expression will be replaced with the
floating-point value.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only If something is
true: * as far as we know - it's science *for certain - it's
mathematics *unquestionably - it's religion
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
On Mon, May 5, 2025 at 01:03 PM, Per wrote:
I want to do something like add MOS=IPB107N20N3 in the PARAMS section on the Navigate/Edit Schematic Block but it moans at me about not being able to resolve the parameter.
Then it appears you know how to add .PARAMs to your symbol already.? That wasn't clear, because the uploaded symbol does not have any parameters added to it yet.
?
However, if you are using anything before the latest versions of LTspice, parameter values must be numeric only, so attempting to do a parameter assignment "MOS=IPB107N20N3" will fail, unless IPB107N20N3 is another parameter whose value is numeric.? The rest of this message is for people who are not running the latest version of LTspice.
?
You can still use PARAMs to change or set the model of a transistor, by defining the transistor models with numeric names.? For example:
?
.SUBCKT IPB107N20N3_L0 drain gate source ...
...
.ENDS
...
.SUBCKT AnotherFET drain gate source
...
.ENDS
?
.model 1 AKO: IPB107N20N3_L0
.model 2 AKO: AnotherFET
.model 3 AKO: ...
...
.PARAM MOS=1
?
But don't forget that both NMOS symbols inside your "bimos.asc" schematic must have their Prefix values changed from "MN" to "X", to make them work with that Infineon SPICE model which is a .SUBCKT model.? Also, if you use the Infineon model named just "IPB107N20N3", that model requires a special MOSFET symbol with 5 pins.? I hope you are aware of how those models are supposed to work.
?
Andy
?
?
|
On Mon, May 5, 2025 at 03:30 PM, John Woodgate wrote:
I've never heard of that procedure. Let's see what others say.
It was OK.? But it probably was not clear exactly what was going on, at first glance.? We are so accustomed to seeing only top-level schematics that we forget there are other kinds, in LTspice-land.
?
Per's .ASC schematic file is a lower-level hierarchical schematic.? It is not meant to be run by itself.? It is meant to be "called" by a higher-level schematic.
?
The .ASY symbol file that accompanies it is the symbol that represents that lower-level schematic.? Symbols for lower-level blocks are supposed to have no Attributes, no simulation commands, and maybe no power supplies.? In that regard, everything appears to have been done correctly.
?
?
?
|
On Mon, May 5, 2025 at 05:38 PM, John Woodgate wrote:
Well, I now have 24.1.8, but the Help still says: The enclosed expression will be replaced with the floating-point value.
It would not surprise me if they did not catch every instance in the Help that needed to be changed.
?
You ought to point out where you saw that in the Help.
?
Andy
?
?
|
Well, it's hardly hidden away. It's the third
short paragraph in the .PARAM help:
To invoke parameter substitution and
expression evaluation, enclose the expression in curly braces.
The enclosed expression will be replaced with the
floating-point value.
On 2025-05-05 23:00, Andy I via
groups.io wrote:
On Mon, May 5, 2025 at 05:38 PM, John Woodgate wrote:
Well, I now have 24.1.8,
but the Help still says: The
enclosed expression will be replaced with the
floating-point value.
It would not surprise me if they did not catch every instance in
the Help that needed to be changed.
?
You ought to point out where you saw that in the Help.
?
Andy
?
?
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
On Mon, May 5, 2025 at 06:22 PM, John Woodgate wrote:
Well, it's hardly hidden away. It's the third short paragraph in the .PARAM help:
To invoke parameter substitution and expression evaluation, enclose the expression in curly braces. The enclosed expression will be replaced with the floating-point value.
There is some chance that statement is still correct as it stands, though maybe confusing.
?
Evaluating a parameter's value from its formula would apply only to parameters and formulas that can evaluate to numbers.? If a parameter holds a text string, then I think it does not need to be evaluated because the string itself IS the value.? Enclosing such a parameter inside curly braces feels like it is unnecessary and should not be done.
?
OTOH, if you can build expressions that use non-numerical parameters (e.g., ModelName+".sub"), then curly braces might be needed, and then that Help statement should be modified.
?
?
?
|
Very possibly. I think that the new feature of
.PARAM, to pass strings, needs to be clearly explained in the
.PARAM Help, as, for example:
?.PARAM can be used in two ways:
1. To pass a numerical value
[Some text]
2. To pass a string
[Some text]
On 2025-05-05 23:38, Andy I via
groups.io wrote:
On Mon, May 5, 2025 at 06:22 PM, John Woodgate wrote:
Well, it's hardly hidden
away. It's the third short paragraph in the .PARAM help:
To invoke parameter substitution
and expression evaluation, enclose the expression in
curly braces. The enclosed expression will be replaced
with the floating-point value.
There is some chance that statement is still correct as it
stands, though maybe confusing.
?
Evaluating a parameter's value from its formula would apply
only to parameters and formulas that can evaluate to numbers.?
If a parameter holds a text string, then I think it does not
need to be evaluated because the string itself IS the value.?
Enclosing such a parameter inside curly braces feels like it is
unnecessary and should not be done.
?
OTOH, if you can build expressions that use non-numerical
parameters (e.g., ModelName+".sub"), then curly braces might be
needed, and then that Help statement should be modified.
?
?
?
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
That's not the 24.1.8 help, then. You must have looked at an older version.
?
On Tue, May 6, 2025 at 12:22 AM, John Woodgate wrote:
Well, it's hardly hidden away. It's the third short paragraph in the .PARAM help:
To invoke parameter substitution and expression evaluation, enclose the expression in curly braces. The enclosed expression will be replaced with the floating-point value.
|
You are right. 24.1.8 has not installed
itself.
On 2025-05-06 07:17, Mathias Born via
groups.io wrote:
That's not the 24.1.8 help, then. You must have looked at an
older version.
?
On Tue, May 6, 2025 at 12:22 AM, John Woodgate wrote:
Well, it's hardly hidden
away. It's the third short paragraph in the .PARAM help:
To invoke parameter substitution
and expression evaluation, enclose the expression in
curly braces. The enclosed expression will be replaced
with the floating-point value.
--
Best wishes John Woodgate RAYLEIGH Essex OOO-Own Opinions Only
If something is true: * as far as we know - it's science *for
certain - it's mathematics *unquestionably - it's religion
|
Apologies for setting things off and thanks for the additional replies. I upgraded to 24.1.8 and now it is working. I did have {MOS} set as the mosfet type for M1 and M2 in my subcircuit so it just ran after the upgrade with no complaints. I haven't checked it out fully but fingers crossed. Not sure about the 5 pin mosfet model for the Si7336ADP. It is listed in the 3 pin symbol mosfet picker.
|
Ooops. Still not having it. I shall sleep on things and try again tomorrow.
|
John Woodgate
Bell, Dave