iterated instances with ipar

[Ralf Geiger]

Hello *,

I like to have a symbol with the parameter
NS (number of serial devices) and under that
symbol is a schematic with an iterated instance.
I like to do now a " Q<0:ipar("NS")> " but I get
an syntax error.

Is this in any form possible or not to do?

Cheers
Ralf

--
no signature

 

[S. Badel]

I do not think you can do this, because you are actually modifying
the topology, not just a parameter value.
however, you probably can easily do this with a schematic p-cell.
I never used schematic p-cells, but Andrew Beckett posted an
example skill code a while ago in this newsgroup, that
does more or less the thing you're looking for.
the link is

http://groups.google.ch/groups?hl=fr&lr=&selm=l2isg0974sdf62g0toufivcqmfrao132ck%404ax.com

hope this helps,

stephane


Ralf Geiger wrote:

Hello *,

I like to have a symbol with the parameter
NS (number of serial devices) and under that
symbol is a schematic with an iterated instance.
I like to do now a " Q<0:ipar("NS")> " but I get
an syntax error.

Is this in any form possible or not to do?

Cheers
Ralf

 

[Ralf Geiger]

S. Badel schrieb:

I do not think you can do this, because you are actually modifying
the topology, not just a parameter value.
however, you probably can easily do this with a schematic p-cell.
I never used schematic p-cells, but Andrew Beckett posted an
example skill code a while ago in this newsgroup, that
does more or less the thing you're looking for.
the link is

http://groups.google.ch/groups?hl=fr&lr=&selm=l2isg0974sdf62g0toufivcqmfrao132ck%404ax.com

Thanks I will have a look at this one.

Cheers
Ralf

 

[Ralf Geiger]

S. Badel schrieb:

that would make parallel devices, not serial.
maybe some simulators support a similar parameter for serial devices,
though.

stéphane

stephane you are right ...

I had the problem to wire 10 Mos Transistors in "series"
with the gate Parallel.

I solved that with iterated instance in a schematic an a
"iterated" pseudo bus for D and S pins. Works good, even
if it looks strange in the first place.

Cheers
Ralf

 

[fogh]

Ralf,

The device you instance, a bipolar I suppose, doesn t it have a multiplicity
parameter ? Maybe called area multiplier , number of emitters, or stg in that trend.

S. Badel wrote:

I do not think you can do this, because you are actually modifying
the topology, not just a parameter value.
however, you probably can easily do this with a schematic p-cell.
I never used schematic p-cells, but Andrew Beckett posted an
example skill code a while ago in this newsgroup, that
does more or less the thing you're looking for.
the link is

http://groups.google.ch/groups?hl=fr&lr=&selm=l2isg0974sdf62g0toufivcqmfrao132ck%404ax.com


hope this helps,

stephane


Ralf Geiger wrote:

Hello *,

I like to have a symbol with the parameter
NS (number of serial devices) and under that
symbol is a schematic with an iterated instance.
I like to do now a " Q<0:ipar("NS")> " but I get
an syntax error.
Is this in any form possible or not to do?

Cheers
Ralf

 

[S. Badel]

that would make parallel devices, not serial.
maybe some simulators support a similar parameter for serial devices,
though.

stéphane

fogh wrote:

Ralf,

The device you instance, a bipolar I suppose, doesn t it have a
multiplicity parameter ? Maybe called area multiplier , number of
emitters, or stg in that trend.

S. Badel wrote:

I do not think you can do this, because you are actually modifying
the topology, not just a parameter value.
however, you probably can easily do this with a schematic p-cell.
I never used schematic p-cells, but Andrew Beckett posted an
example skill code a while ago in this newsgroup, that
does more or less the thing you're looking for.
the link is

http://groups.google.ch/groups?hl=fr&lr=&selm=l2isg0974sdf62g0toufivcqmfrao132ck%404ax.com


hope this helps,

stephane


Ralf Geiger wrote:

Hello *,

I like to have a symbol with the parameter
NS (number of serial devices) and under that
symbol is a schematic with an iterated instance.
I like to do now a " Q<0:ipar("NS")> " but I get
an syntax error.
Is this in any form possible or not to do?

Cheers
Ralf

 

[fogh]

Ah, sorry, I did not pay attention to that bit. I saw some DKs that support a
"number of series elements" parameter, but they tend to have mistakes. Last
time, the parameter was handled properly in pcells and in te main simulators but
is forgotten in some siminfo (au*). And IMHO verification is one thing a DK
author should not screw up.

The bus notation works nicely for series connections, but you have to expand
the ipar because only values are accepted inside those <bra:ck:ets>

Please send your schematic pcell when you have it right.

S. Badel wrote:

that would make parallel devices, not serial.
maybe some simulators support a similar parameter for serial devices,
though.

stéphane

fogh wrote:

Ralf,

The device you instance, a bipolar I suppose, doesn t it have a
multiplicity parameter ? Maybe called area multiplier , number of
emitters, or stg in that trend.

S. Badel wrote:

I do not think you can do this, because you are actually modifying
the topology, not just a parameter value.
however, you probably can easily do this with a schematic p-cell.
I never used schematic p-cells, but Andrew Beckett posted an
example skill code a while ago in this newsgroup, that
does more or less the thing you're looking for.
the link is

http://groups.google.ch/groups?hl=fr&lr=&selm=l2isg0974sdf62g0toufivcqmfrao132ck%404ax.com


hope this helps,

stephane


Ralf Geiger wrote:

Hello *,

I like to have a symbol with the parameter
NS (number of serial devices) and under that
symbol is a schematic with an iterated instance.
I like to do now a " Q<0:ipar("NS")> " but I get
an syntax error.
Is this in any form possible or not to do?

Cheers
Ralf