I guess every respecting itself book on electronics considering the astable multivibrator including its working condition wave forms. The multivibrator in question is to work on 400Hz. My simulation proved it does not work as expected. May be I missed some during sim or even did something wrong. May someone check it out, please? Just in case power supply 10V.
multivibrator.7z (814.7 KB) this is bad one
multivibrator.7z (814.9 KB) no ground is bad too
multivibrator.7z (815.2 KB)
Looking at your schematic:
- 1V power is a bit low for this circuit. I’m not sure it can even work at such a low voltage.
- In this configuration, it works with NPN BJT’s, not with PNP.
- For the circuit to work with PNP, you’d have to reverse the polarity of the voltage source (Better: Flip the whole circuit so voltages go “from top to bottom”.
- There is no GND reference, this is mandatory for spice simulations.
- Printing texts over each others makes my neck hairs rise.
- Entering resistor and capacitor values as you did, may work, but it’s at least overly complicated.
Apart from the issues above, one of the most common faults is a mismatch between the spice models and the schematic symbols, but I did not even look at that because of those other issues.
I suggest you start with some working examples. There are several threads on this forum with simulation examples. The thread below for example also has an example of this classic astable multivibrator.
Or a direct search, which finds a bunch of threads here:
https://forum.kicad.info/search?q=kicad%20simulation%20examples
You seem to not live in germanium pnp times. Those times at all schematics GND (+) was at bottom and at top of schematic you had -.
Everyone is used to understand this multivibrator with transistors at bottom.
The BC556 datasheet included in the zip file has a Vbe of 600mV, so it’s not Germanium. And before that “Better” I wrote:
The way it is drawn the currents a flowing in the wrong direction through the BJT’s and it won’t work that way.
The third version of my multivibrator is correct as of polarity and ngspice requirements. Transistors, what to say, cannot help it.
This 3rd version is working. The only thing missing is that you have to add what simulations you want to run. Go to: Schematic Editor / Inspect / Simulation. Then In the Simulation menu add a new simulation tab for a transient simulation:
Thank you very much, indeed!
From your mistakes, I deduce you did not follow any sort of tutorial but were just experimenting. Reading some tutorials can be more efficient. There is a quite nice written tutorial on the ngSpice website
One day I learned that it is sometimes impossible to get a hand on a ngspice required part model at all. And then I got busy with different things but, these days, I am back again for electronics, unexpectedly. I do not want to leave it but circumstances. Some manufactures has their own ngspice and all of their product models are made for their own sim software. And that is why to dive into KiCAD is a little questionable quest after all.
I cant turn your statement around just as easily. I am quite annoyed by all the walled gardens and attempts to increase your own market share by blocking copetititors. That is not the way that fair competition is supposed to work. But unfortunately there is very little fairness in this world. And still I believe that fairness and helping each other is the only long term solution. And that is a big part of why I use open source software such as KiCad and ngSpice, don’t even consider using any other spice simulator, even if it’s “better” in some sort of niche. If I can’t simulate some fancy new IC because there is no model available with ngSpice, then I probably simply just won’t use that IC. For most stuff, there are plenty of alternatives to choose from, unless you get into the really “high end” modern stuff.
1 Like
For models, always go to the manufaturers first. For very old devices, it can be a problem finding models at all.
Whether the manufacturers’ models are SPICE, PSPICE or LTSPICE doesn’t matter, ngspice will work with those as well.
I thought of it on the contrary. I wanted to do models myself on datasheet bases. Oops!
OK. There is a thyristor that supposed to be on/off on 1000Hz frequency with 120A current flow through it. Thyristor type T161 is suitable for the purpose. You think KiCAD quite reliable for it, do you? The idea of using the thyristor for the purpose is questionable in the first place. Am I here yet?
What sort of models?
ngSpice has a whole lot of built in models for BJT’s FET’s, Opamps, I think you can add parasitic parameters to passive parts and a whole lot more. (I don’t use ngSpice very much and never really investigated this part).
You can work with, (and design your own if you wish) subcircuits or IBIS models. Looking around on Ngspice, the open source Spice circuit simulator - Special features I also see other things such as mixed mode simulation and combining uC C code with analog electronics.
ngSpice is not the latest and greatest in the simulation world, but it is a capable simulator, it’s actively maintained, and it’s well integrated in KiCad. And that is enough for my own purposes.
Then you are lucky person after all!
1 Like
Seems to be a response to me, but I don’t understand it at all. Sorry.