Hi.
I want to use MMBTA42 in my Ltspice simulation for which I turned to internet to get the spice model.
I got two models form different places

.MODEL DI_MMBTA42 NPN (IS=51.0f NF=1.00 BF=194 VAF=312 + IKF=0.182 ISE=34.9p NE=2.00 BR=4.00 NR=1.00 + VAR=24.0 IKR=0.450 RE=11.6 RB=46.3 RC=4.63 + XTB=1.5 CJE=58.7p VJE=1.10 MJE=0.500 CJC=18.9p VJC=0.300 + MJC=0.300 TF=481p TR=115n EG=1.12 )

( https://www.diodes.com/spice/download/4448/MMBTA42Q.spice.txt )

.model MMBTA42 NPN (Is=34.9f Xti=3 Eg=1.11 Vaf=100 Bf=2.65K Ne=1.708 Ise=16.32p Ikf=23.79m Xtb=1.5 Br=9.769 Nc=2 Isc=0 Ikr=0 Rc=7 Cjc=14.23p Mjc=.5489 Vjc=.75 Fc=.5 Cje=49.62p Mje=.4136 Vje=.75 Tr=934.3p Tf=1.69n Itf=5 Vtf=20 Xtf=150 Rb=10)
(In datsheet attched)

Could anyone please help me identify which one is correct. I have attached datasheet for the reference as well.

This is not the only model, I'm getting different models for different components as well..

 

Both may be "correct", since both are only approximations to the real thing. One or other might be preferred, depending on which behaviour/properties of the component are of particular importance in your simulation.
Try both in your sim and compare results.
In general, the more accurate/refined the model is, the slower the sim.

 

Both may be "correct", since both are only approximations to the real thing. One or other might be preferred, depending on which behaviour/properties of the component are of particular importance in your simulation.
Try both in your sim and compare results.
In general, the more accurate/refined the model is, the slower the sim.

I tried to simulate both of them in my circuit. Vb comes around 4.3 V with first model and around 800 mV using second model.

How to know thi which one to use if I have only part number given in the original circuit?

 

This is not an uncommon occurrence. I have over a dozen models for the TIP31C power transistor and most of them are unusable. There is no convenient way to judge or guess which models may be appropriate for a given purpose. One tedious process is to enter the parameters of multiple models into a spreadsheet in a standard form so that individual parameters can be compared side by side. If you don't understand the purpose of one or more parameters, then this exercise is not likely to produce satisfactory results.

If they ever establish a central authority for cataloging and maintaining models it would go a long way towards resolving the problem, but such a situation is extremely unlikely IMHO.