RB & rbb'
I'm looking at Noise performance for this & other circuits.
For good LN transistors, noise can be modelled simply with rbb'
From what I can see, RB = rbb' in SPICE
I can get a good estimate of rbb' from noise contours to replace the default RB = 10R which is far too optimistic for audio noise.
What are the other implications and are they small, eg VHF & UHF response, if I just change RB?
I note the default LTspice BC547s have superb RB=1R
. Wish I knew where they bought their stuff.
But even Cordell's stuff is suspect. His npn 2n4401C has RB=13 while the pnp 2n4403C has RB=38 which is more believable and closer to Bcarso and my own Jurassic measurements. I've always found npn to have higher rbb'
I get rbb' from the high current 1kHz noise contours and that's been pretty good for me.
I know rbb' at 100MHz may be different but if I just change RB, to get closer to real life audio noise, will I be mucking up my ability to see 100MHz oscillation?