LTSpice MESFET models

Just plugged it in and LTSpice (rightfully ...) complains about all the missing data. There is nothing behind Xti, N, Eg and so on.

n

could try to delete the empty parameters, can't very well be important if they are empty

-Lasse

Thanks, gents. I did that already, as I should have mentioned. Even with default parameters only, it doesn't seem to know about anything resembling NMESFETM1 or STATZ_MESFETM1.

I saw what looked like a SPICE model for the ATF34143 (the big brother of my device), which called out a device "GAASFET", but LTSpice doesn't like that either. Using the NMF model, (which seems to be the vanilla SPICE way of doing a MESFET) a simple simulation sort of runs, but just about all the named parameters get rejected.

Cheers

Phil Hobbs

One option would be to take a similar FET that LTSpice does eat and compare the items line by line, parameter by parameter. If you get totally stuck the best forum for this is the Yahoo Groups LTSpice group, and there mainly Helmut Sennewald.

Am 08.05.2012 20:19, schrieb Phil Hobbs:

in is a cross reference. Spice2/3 and ADS Statz don't seem to have much in common.

Agilent ADS, Genesys, Microwave Office and some others support the Statz model. The handbooks apear to be on the net, so that should be a place to search.

regards, Gerhard

Normally it does, but I don't know about LTSpice:

This App Note could help:

[...]

It exists in PSpice as GaAsFET, Statz Model, Level=2. It's probably there in LTspice somewhere, Mike is pretty thorough about covering everything. ...Jim Thompson

Okay, after I gave up and resorted to Reading The Fabulous Manual, it sort-of works. A bunch of the parameters are missing from the LTspice version, but it claims to be based on the Statz model also, so it may be better than the behavioural version.

Quite ridiculously, it leaves out the flicker noise parameters.

Thanks, all.

Cheers

Phil Hobbs

Phil,

LTspice apparently addresses MESFET/GaAsFET, but no models (For a possible model see below):

From LTspice Changelog:

02/07/04 Fixed a problem evaluating the MESFET saturation region. 02/17/02 Corrected some errors in MESFET current monitoring

From mesfet.asy:

Version 4 SymbolType CELL LINE Normal 16 12 16 84 LINE Normal 48 80 48 96 LINE Normal 48 16 48 0 LINE Normal 16 16 48 16 LINE Normal 0 80 16 80 LINE Normal 16 80 48 80 WINDOW 0 56 32 Left 0 WINDOW 3 56 72 Left 0 SYMATTR Value NMF SYMATTR Prefix Z SYMATTR Description GaAs MESFET transistor Note: You must supply the .model for this device PIN 48 0 NONE 0 PINATTR PinName D PINATTR SpiceOrder 1 PIN 0 80 NONE 0 PINATTR PinName G PINATTR SpiceOrder 2 PIN 48 96 NONE 0 PINATTR PinName S PINATTR SpiceOrder 3

Possibly useful link (maybe):

Here's an NMOS depletion-mode model (Fairchild, BSIM) which might SCALE to match your needs:

***********************************************************************

• HIGH VOLTAGE DEPLETION NMOS TYPICAL MODEL CARD *

*********************************************************************** .MODEL NDP NMOS LEVEL = 6 +VTH0 = -1.7178513 K1 = 0.9981337 K2 = -0.0681299 +K3 = 37.6482304 K3B = -3.53199 W0 = -1.641794E-7 +NLX = 1E-9 DVT0 = 0.1073743 DVT1 = 3.534263E-3 +DVT2 = 0.0364547 DL = 3.2E-7 DW = 2.367287E-7 +U0 = 570.3449457 UA = 0 UB = 4E-18 +UC = 4.7804E-3 VSAT = 1.02031E7 A0 = 0.6869448 +KETA = -8.8323E-3 A1 = 1 A2 = 1 +RDS0 = 24.3914569 RDSW = 1.76728E3 VOFF = -0.06 +NFACTOR = 0.5281248 CDSC = 2.001847E-5 CDSCB = 0 +ETA0 = 1.063091E-3 ETAB = -0.036365 DSUB = 0.2415699 +PCLM = 13 PDIBL1 = 0.01 PDIBL2 = 1E-5 +DROUT = 6.3088897 PSCBE1 = 1.040507E9 PSCBE2 = 1.003176E-7 +PVAG = 6.0679758 UTE = -1.3 KT1 = -0.32 +KT1L = 0 KT2 = 0.015 UA1 = 4.422664E-9 +UB1 = -5.74133E-18 UC1 = -0.056 AT = 3.3E4 +TOX = 3.13E-8 XJ = 4E-7 NPEAK = 1.7E17 +NSUB = 2E14 SUBTHMOD= 1 SATMOD = 2 +BULKMOD = 1 XPART = 1 XT = 1.55E-7 +VBM = -5 ETA = 1 VGHIGH = 0.15 +VGLOW = -0.15 CIT = 0 JS = 5.325E-5 +RSH = 0 CGDO = 6.383E-10 CGSO = 6.383E-10 +CGBO = 1E-10 CJ = 2.818288E-4 PB = 0.88 +MJ = 0.5735524 CJSW = 5.415684E-10 PBSW = 0.65 +MJSW = 0.268275

HTH! ...Jim Thompson

Thanks. Starting from where I am, scaling that model would be pretty time-consuming. However, Reading the Fabulous Manual probably got me there--I'm just putting the package parasitics in.

Cheers

Phil Hobbs

Maybe not, Just varying M and VTH0 works quite often for me.

Have fun ;-) ...Jim Thompson