Re: fast NPN in LT Spice

On 6/06/2024 1:46 pm, legg wrote:

On Tue, 4 Jun 2024 00:10:32 +1000, Bill Sloman <bill.sloman@ieee.org>
wrote:
<snip>

Gosh, what a hideous mess, in many respects.

>
Do tell us why. You do claim to revel in electronic discussion.

 Perhaps its the nonlinearity of the output stage, which is biased off.
 So it's a switch, but the slow speed of the output, if engaged,
results in a stretched pulse.

The pulse stretching happens between the two BFR92 transistors, which act as a classical emitted-coupled monostable, which stretch the input spike out to about 30nsec
Q3 and Q4 just level shift this down to provide an ECL-level output pulse which is also 30nsec wide.
There are other ways of doing this kind of level shifting and you can use faster transistors.

The assertion that spice parameter Tf is related to spec sheet Ft is
only a guess.

Not one that I made. The BFR92a is just the default broad-band transistor, which is why Jeroen Belleman picked it too.

The bfr92a model written into your simulation turns out to
be part of a more complete model published as a die-within-a-package.
There's little difference in performance when substituted into the
simulation.
 If all the models with Tf<20ps are evaluated, you get unpredictable
results. Note that the bfr92a model doesn't actually meet this
limitation, but other similar models do (~bfr93). There are roughly
270 of them.
 Each will either:
 - fail to engage with the slow output detector.(31)

If they aren't fast enough to get triggered as emitter-coupled monostable   by a pulse that is 400psec wide at hallf-maximium voltage of 100mV.

- act roughly like the original simulation.(217)
- oscillate at an unrelated frequency.(19)

Broad-band transistors will oscillate without a base-stopper of adequate resistance - I tended to end up with resistors between 22R and 33R. There are better solutions, but in the work I did it wasn't worth the trouble of finding them.

- stall.(1)

Up the gain.

- give incoherent wild results (2)

Change the base-stopper resistance.

http://ve3ute.ca/query/Tf_20ps_vs_bfr92a.zip
 Just why one model does one thing, while another does something
else might be interesting to figure out.

No model is perfect. If you model something and it seems to work, it's worth putting together a real circuit (which takes longer) and seeing what it actually does.
If it doesn't work, or doesn't work all that well. fiddling with the model may point the way to something that might work better.
--
Bill Sloman, Sydney
--
This email has been checked for viruses by Norton antivirus software.
www.norton.com