Hello , i have a netlist model for LTSPICE which i want to implement for building an oscilator using spice text i have trying to import the model shown bellow.
however when i do the check and save on spice text editor it marks red the .model line as if its not fitting the standart of some sort, maybe the PCHAN parameter which is level 18 parameter.
where did i go wrong importing this model?
what kinD of cadence virtuoso spice syntax could fit this model if there is a mismatch?
.SUBCKT SMS2001-C d g sM1 d g s s XSMS2001-CRg g s 1GDb d s dbody.model XSMS2001-C VDMOS(KP=4.5359 RS=0.0253 RD=0.1143 RG=60.83 VTO=-0.73+LAMBDA=0.001 CGDMAX=165p CGDMIN=23p CGS=166p TT=720n a=0.77+m=0.1 Vj=0.67 Cjo=10.0pF PCHAN).model dbody d ron=0.1 vrev= 20 vfwd=2.ends
You can't do this (I'm not sure why you would want to either, since spectre is normally used for IC design, and this is presumably a discrete component?).
The issue is that VDMOS is a proprietary model in LTSPICE and not available in other simulators (so it's not recognised in Spectre). See http://ltwiki.org/LTspiceHelp/LTspiceHelp/MOSFET_Model.htm
Hello Andrew, i am trying to use this specific model in cadence to use it as oscillators that i implemented before in gpdk or umc90nm
Bellow you can see that i interpreting each parameter from the model for its meaning.Where can i find a MOSFET netlist syntax that would be the best fit for this model which cadence virtuoso would accept?
I don't really see why you'd use a model from a completely different technology in another process. You're not going to be able to implement it (although gpdk is a fake process anyway, you might as well use a transistor from that process rather than inventing another one).
You might be able to use mos1, mos2 or mos3 (see "spectre -h mos1") but this doesn't support the same parameters.
However, this seems to me an utterly pointless exercise. I can't tell you how to adapt a proprietary model into an equivalent model - if you really wanted it to match, you'd have to re-characterise it. But doing this seems bizarre to me as unless this corresponds to something that is consistent with the process you're using, it's a waste of time.
Unless I'm missing something...
Hello Andrew , so you are saying that this netlist model doesnt fit for being used in buillding an oscilator .
How can i recognise that a mosfet model that if front of me could be used for such purposes?
Put simply, if you're trying to create a design on a particular technology - be it a real process such as UMC90nm or an artificial process intended for demonstration such as gpdk090 or gpdk045 - then it doesn't make sense to introduce transistors into your design which are not representative of devices that would exist in that technology. Given the name of the model and a bit of googling, this appears to be a discrete p-channel power MOSFET device, so I'm not sure what value there would be in using that model as part of an IC design - it's utterly meaningless.
Now there might be a situation where you were driving a signal off-chip to connect to this off-chip transistor, but that doesn't appear to be what you're asking for here. If you were, I would ask the transistor vendor for an alternative model (e.g. a PSPICE model) which could be used in spectre. However, I don't think that's what you're doing here - I'm guessing you've found some oscillator design using this particular transistor (which would seem a rather strange use of a power MOSFET, to be honest) and are trying to re-use it in your design?
Anyway, people design oscillators (with all sorts of different architectures and types) on IC processes all the time - so there's no need to have a special type of transistor to design an oscillator.
I would suggest you speak to your supervisor - hopefully they can give you some practical guidance of doing real-life IC design.