I have a simple sub-circuit design that I want to test in LTspice. I've already done the basic simulation and it behaves as expected, but what I specifically want to look at now is how much the Beta (hFE?) of the transistor impacts circuit performance. I'd love to just swap in a few transistors at the extreme ends of the spectrum, but I have no idea which BJT models have especially high or low Beta. I only have the standard parts that come with LTspice, because I run sims on a lot of different computers and I got tired of trying to keep them all sync'd up in terms of which computers have which extra parts libraries.

So, I see three possible solutions to this:

1. I simply try every single transistor and keep detailed records on all of them to identify good candidates at both extremes. This sounds horribly tedious and I'm hoping to avoid it.
2. Someone here tells me what one or two examples at both ends of the spectrum are, within the standard LTspice parts list.
3. Someone here tells me how to identify Beta from within the numbers that define each BJT model (is there even a way to do that, or is the Beta a function of a whole lot of separate variables in the Spice definition?)

So, if anyone here can advise me on some standard LTspice BJTs with unusually high or low Beta, or can tell me how to find that in the Spice parameters, I'd really appreciate it. Thanks in advance for your help!

 

Hello

The native BJT models provided with LTspice are modeled per the manufacturer's specifications. The manufacturer is shown in the "pick list" when you select a transistor. The hFE data should be in the corresponding datasheets. The "Ideal" models, NPN and PNP, are exactly that...ideal, so I wouldn't use them for what your wanting to do.

The HFE parameter is BF. You could step this parameter while monitoring a IV curve until you get the extremes your looking for..

eT

 

as to my knowledge you can't "step" the BF -- instead you need to "step" it manually as ::
↓▼↓ for "SPICE Directive" :
.model MYNCLONE ako:2N5769 NPN(BF=371 mfg=Clone)
+ set [ Ctrl ]+ [ Mouse R-click ] on NPN Attribute Value to MYNCLONE
or in .SUBCKT (.CIR) -- definition (file)
* a default / (HF/"near-ideal") NPN definition
.MODEL QX NPN(BF=333.333)
↑▲↑

 

Yes, you can step BF.
In the transistor model definition you can replace 'BF=100' (for example) with 'BF={beta}' and then add a .step param beta ...... command.

 

You will find it useful to see my examples and my extended libraries.
http://bordodynov.ltwiki.org/

 

The HFE parameter is BF. You could step this parameter while monitoring a IV curve until you get the extremes your looking for..

eT

Perfect, thanks so much! I'll try that this afternoon.

 

I tried both approaches - simply searching through existing models based on BF parameter, and also stepping the BF value. Both work perfectly for my needs. Thanks for the help!