Possible bug

This exact differential amplifier with current mirror load circuit works just fine* on the circuit simulator but on SimulIDE 4.15 either gives a Circuit ERROR, or if I can get it to run then when the left side input is lower than the right side then the output is just slightly lower than the Vcc instead of 0 or close to it.

*When 3 volts is disconnected the output is a few mV instead of 0, so not perfect, but close enough.

PNP transistors do not work in the simulator. I checked the operation of the keys on the PNP transistors and the multivibrator. In the versions of the program 0.4.15 and 0.5.16 there are errors when working with these transistors.

Fizik_S Fizik_S wrote:PNP transistors do not work in the simulator.

That explains a lot, thanks.
And I'll check out the zip file you posed.

Yes.. there is a problem here.
I will have a look.

Thanks for reporting.

This issue should be solved at trunk Rev 587 and 0.4.15 Rev 242.

You can test it right now, new stable version released:
https://www.simulide.com/2021/09/simulide-0415-stable-released.html

PNP transistors work a lot better now, but I notice they sometimes keep a charge even if voltage is turned off, almost like a capacitor.

PNP transistors work a lot better now, but I notice they sometimes keep a charge even if voltage is turned off, almost like a capacitor.

Can you elaborate a bit more please...
where exactly is that "charge" and what exactly do you mean by "voltage is turned off".

Sorry for being so vague.  I mean turning off the voltage to the base.  Band by turning off the voltage, I mean pressing the button on the Fixed Voltage component.
This only seems to happen when transistors are connected to each other and after clicking the voltage on and off a few times.  For example, in the circuits posted above by Fizik_S clicking the voltage on and off a few times to the base of each PNP always gives the desired results.  But on the circuit I posted, with the 5V and 2V on, for a few clicks of turning the 3V on and off it behaves as expected, but after a few clicks the probe stays stuck a certain voltage (not always the same) and it takes turning the 5V off and on again to clear it.

Thanks, now I undestand.

And yes... starts working ok, but at some point it goes crazy.
I will have a look.

Finally I found the problem was in the Base-Collector direc connection in Q6.
This doesn't work in simulide and It's not easy to solve.

By now a workaround is using a very small resistance (like 1 mΩ) instead of direct connection.
Something like this seems to work:

Thank you very much.  That's a very big help.