low voltage to drive LED flash (not successful in simulation, but in breadboard successful)

[ankyxia06]

the same as yours but no oscillation and no LED blink.  it is weird

[tipa]

Of course, I will send the file. If you look closely, then asterisks appeared in the names of the transistors, which means that there have been some changes in the model. I changed only one parameter for both transistors. This affects the 'heel' on the characteristic. (You can compare the input and output on the curve graph)
And in the simulation set conditions equal to zero
And your program has nothing to do with it. It's all about setting up models and analysis...
Good luck! Thanks for the chat.

Транзисторы стали похожи на германиевые, с меньшим напряжением на P-N переходах

[tipa]

I also worked a little on the circuit, optimized the parameters. We eat from 0.94 volts! And we set fire to three LEDs!

 

[tipa]

And this is a present model of an ultra-sensitive LED for you (maybe someday they will create it in hardware)

 

ут...)

[ankyxia06]

Hi Tipa,

amazing , it works.

this time I did not change to set to zero under initial condition but I did change model on transport saturation current and it still worked. so I wonder. 

1,  how does initial condition impact this circuit, does it mean all capacitor etc are set to zero by choosing this?

2,  I know saturation current, but I don't know why it has to be this large 1e-09??

appreciate you can detail on this or do you have any links on this if you hate to type so many words.😂 

[tipa]

Hi, darling! It's good that questions arise - it means the person understands, which means that I explain well. ('When you praise yourself, it is at least sincere', someone smart said)
There is a good site where the forum has several branches of the Multisim topic. I recommend registering there. There is also an English version of it. In the Yandex browser INSTANT TRANSLATION. You right-click on the selected text, and the translation immediately pops up in a pop-up window.
http://kazus.ru/forums/forumdisplay.php?f=38
There is a smart uncle bordodynov there who will answer all questions if he is in a good mood. Issue questions in the appropriate branch, in any language, they will definitely help.
Sometimes the simulation doesn't want to run, or crashes after a while. This is typical of quartz oscillators. Then a push is used, in the form of a pre-charged capacitor, or a magnetized coil

2,  I know saturation current, but I don't know why it has to be this large 1e-09??
I put it from the bulldozer, somewhere I spied it in some model. It affects the opening threshold of the P-N junction
And you were not interested in how I made models of ultra-sensitive LEDs?

[tipa]

Вы приводили схему на с9012, с9013 транзисторах. Построив модель по их вольт-амперным характеристикам, можно попробовать ещё раз просимулировать, чтобы без мухлежа всё было.
Вот модель с9012 на основе ztx549. Кривые совпадают с даташитовскими

+ (
+ Level=1 IS=2e-15 BF=200 VAF=55 IKF=1.8 ISE=6e-14 NE=1.45 BR=50
+ VAR=8 IKR=0.2 ISC=3.5e-14 NC=1.23 RB=0.7 RE=0.063 RC=0.122 CJE=1.209e-10
+ VJE=1.041 MJE=0.4685 TF=5.1e-10 CJC=5.5e-11 VJC=0.1928 MJC=0.3541
+ TR=3.6e-09
+ )

[tipa]

Вот, построил модели S9012 и S9013. Они почти не отличаются, поэтому не стал заморачиваться, так как долго пришлось подгонять два параметра. 1-влияет на величину тока коллектора на диаграмме характериографа, 2- изменяет угол наклона выходной характеристики. Таким образом получили комплементарную транзисторы с одинаковыми моделями.

[tipa]

Спасибо за оценку! Люблю, когда задают вопросы, на которые я могу ответить. Постараюсь показать на графиках, как влияет IS.

[tipa]

I connected the simulation results into a common graph, for clarity.