The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> High-Power Design >> simulating HV devices in Spectre
https://designers-guide.org/forum/YaBB.pl?num=1132058634

Message started by vivkr on Nov 15^th, 2005, 4:43am

Title: simulating HV devices in Spectre
Post by vivkr on Nov 15^th, 2005, 4:43am

Hi,

This is more of a simulator issue, but specific to HV. I get a message from Spectre telling me that the voltage across the gate oxide exceeds the breakdown limits. However, I probe the voltages and I can see that this is not the case, especially as I have a HV device which is capable of more than 3x-4x times the voltage that Spectre is using for it's test.

I have a safe-operating check running and it shows no problems. I suspect that there is an internal setting in the simulator which is set to around 8 V. Does anyone know how to adjust this?

Thanks
Vivek

Title: Re: simulating HV devices in Spectre
Post by bernd on Nov 15^th, 2005, 9:55am

If you get a Spectre Warning similar to these

Warning from spectre during DC analysis `dcOp'.
Mn: Vgd has exceeded the oxide breakdown voltage of `vbox' = <someValue> V.

then vbox is a bsim3v3 operating region warning control parameter
which defines the oxide breakdown voltage.
It is 1e9*tox, where the value for tox, the gate oxide thickness, is defined in
your bsim3v3 transistor model.

Bernd

Title: Re: simulating HV devices in Spectre
Post by vivkr on Nov 16^th, 2005, 12:55am

Hi Bernd,

Thanks for the tip. I think I can see the problem. The tox parameter refers to the active gate oxide thickness on the source end which is indeed small, but on the drain side, this oxide thickness is considerably large, but this is not modelled in BSIM3 as there is only one tox parameter.

I guess I just have to get used to this warning.

Vivek

Title: Re: simulating HV devices in Spectre
Post by jbdavid on Feb 3^rd, 2006, 12:23am

Or use (or write in Verilog-A) a transitor model that is appropriate for HV devices (bsim3v3 is appropriate for .5u down to about .13u standard CMOS transistors.. but it has no idea of a Variable thickness Oxide.. AFAIK..