Dave,
?
I don't know if you didn't see this before?
?
The BZX84B33VLY is a 33 V zener.? It is very unwise in my opinion to re-define it as a 3.3 V zener.? The power supply's intended output is +/- about 30 V, so that will not work.? It might 'let out the magic smoke'.
?
In your simulation, those zeners are in breakdown!? The regulated outputs exceed 3.3 V, turning ON those "protection" zeners.? That's not right.? (That may be why the simulation crawls.)
?
Not only that, but you created a conflict between LTspice's built-in BZX84B33VLY model and your modified "AKO" model.? Dangerous!
?
Also the reference diodes in the original circuit had a voltage of around 1.24 V, not 0.68 V.
?
Andy
?
