Regarding a Notice from Spectre

[meghna]

Hi,
When I run any simulation (after clicking the option diagnose option to
'yes'), I get the following massage at output log window-

"Notice from spectre at time = 76.2238 ps during transient analysis
`tran'.
Maximum value for quantity `I' has increased to 855.624 mA (detected
at V38).
If this value is unreasonably large, it will result in an overall
loosening of the convergence criteria. In this case, consider performing
transient analysis with errpreset=conservative or
relref=alllocal."

Is this warning harmful or dangerous? (I am getting the correct outputs
without any convergence problems.)

Regards
Meghna

 

[G Vandevalk]

Have you investigated the "event" that caused the current spike?

What were your max I values before this event?

Sometimes you get numerical "ringing" due to some unrealistic (or real!)
event that gives the simulator fits.

It would be prudent to check out where this happened.

If you were doing some power calculations and you think this is a simulator
artifact, it may be wise to ignore this area of time.

It may be very tricky to pin down why (or blatently obvious!)

- G

"meghna" <meghna@protected_id> wrote in message
news:c6d3c59172572c3bb86c417a2289daee@localhost.talkaboutcad.com...

Hi,
When I run any simulation (after clicking the option diagnose option to
'yes'), I get the following massage at output log window-

"Notice from spectre at time = 76.2238 ps during transient analysis
`tran'.
Maximum value for quantity `I' has increased to 855.624 mA (detected
at V38).
If this value is unreasonably large, it will result in an overall
loosening of the convergence criteria. In this case, consider performing
transient analysis with errpreset=conservative or
relref=alllocal."

Is this warning harmful or dangerous? (I am getting the correct outputs
without any convergence problems.)

Regards
Meghna

 

[S. Badel]

"Notice from spectre at time = 76.2238 ps during transient analysis
`tran'.
Maximum value for quantity `I' has increased to 855.624 mA (detected
at V38).

Is this warning harmful or dangerous? (I am getting the correct outputs
without any convergence problems.)
If you do not expect such large current it can be worth investigating. I've usually

had this type of warning in two cases. First one is when there are transistors with
bulk floating (if two transistors have their bulk tied together, but floating, it does not
trigger any warning) . Second is with ideal sources, when set to an unreasonably small rise/fall
time (apparently your warning is at a source node).

stéphane

 

[meghna]

Thanks for your reply. I am using using the ideal sources with small
rise/fall time, so this warning is coming. One more warning, which I
sometimes get
is-
"Zero diagonal found in Jacobian at `I12._284'.
Reordering Jacobian."
What does it mean and how to remove it?
Thanks and Regards

 

[S. Badel]

Thanks for your reply. I am using using the ideal sources with small
rise/fall time, so this warning is coming. One more warning, which I
sometimes get
is-
"Zero diagonal found in Jacobian at `I12._284'.
Reordering Jacobian."
What does it mean and how to remove it?
Thanks and Regards

I think floating nodes can cause this, or a loop of ideal sources (ie, two or more voltage sources
in parallel, or current sources in series).

but i'm not a spectre specialist, just an user...

stéphane

 

[meghna]

I have some floating nodes, which may be problematic. One more notice is--

"Gmin = 1 pS is large enough to noticeably affect the DC solution.
dV(I13.M6:int_s) = 114.185 mV"

Here what does it mean by 'I13.M6:int_s'?
I could not understand ':int_s' part.
Thanks!

 

[Andrew Beckett]

On Thu, 21 Apr 2005 01:07:38 -0400, "meghna" <meghna@protected_id> wrote:

I have some floating nodes, which may be problematic. One more notice is--

"Gmin = 1 pS is large enough to noticeably affect the DC solution.
dV(I13.M6:int_s) = 114.185 mV"

Here what does it mean by 'I13.M6:int_s'?
I could not understand ':int_s' part.
Thanks!

It's the internal source - the other side of the resistor which is part of the
source.

Regards,

Andrew.