Electronics > Beginners

not gate oscillator

(1/1)

yalect:
Hello,
I would like to ask you if the following circuit (not gate oscillator) circuit will work,knowing that I simulated it on software and worked as shown on the picture
Thank you

T3sl4co1l:
Sort of.

What you've simulated is, in part, an artifact of the XSPICE simulation engine.

Multisim (appears to be what you're using) uses XSPICE.

In XSPICE, logic gates are simulated by converting analog inputs into digital signals.  (You can observe digital-only signals on the "oscilloscope", inbetween gates.  The output you're viewing may be a digital version as well; I forget exactly.)

Digital signals are simulated based on events: every time an input changes state, that state is propagated through the logic circuit, according to the propagation delays specified in each part.  Finally, outputs are converted to analog signals where needed.  (You can always force a board-level signal to convert into analog form by introducing a 1m resistor between gates.)

This is very different from how SPICE (the normal analog stuff) is handled, which is by a calculated timestep.  For each instant in time, all the currents and voltages are calculated, and a new step of time is calculated (and refined if the result is in error).

The combination of these two systems can be buggy at times, but that's not applicable here I think.  If nothing else, appreciate that what you are seeing here is in no way representative of any real logic IC.  Sort of...


So, the "gotcha" is this:

A real circuit may simply not oscillate at all (e.g., CD4069UBE), and output a constant voltage approximately VDD/2.

Or it may oscillate, but with a very low amplitude, because the output (which has a typical resistance in the 30-1000 ohm range, depending on family) is heavily loaded by the 200 ohm resistor, and that in turn makes a small change on the 1uF capacitor, which is amplified by the transistors inside the chip to give what output is seen.

Or it may simply work as shown, with an amplitude (on the capacitor) determined by the input threshold (e.g., 74AC14 has input hysteresis, so will run with something like 300mVp-p there, and has a strong enough output (~30 ohms equivalent) to drive the 200 ohm resistor).

So it depends.

You most certainly will not observe the first case, using a digital (code) model: you will note, because the digital simulation is event driven, there cannot possibly be a simulation of analog feedback, so the output will necessarily dither, perhaps at impossible frequencies (try reducing the capacitor to 1pF!).

I also have no idea what logic gate is being simulated there, if it has realistic propagation delays (a few nanoseconds, say?) or if it's "ideal".

If nothing else, perhaps the best lesson I can give, is to note that a simulator is just that, a simulator -- a pile of code on your machine.  It doesn't know anything about the real world.  Its designers know some things, and provide some models that are more or less realistic, but they also provide primitives that work with the simulator but which may not have any meaningful equivalent in reality.

HTH,
Tim

yalect:
thank you for your reply
but most not gate oscillator circuit works with two gate, one is SCHMITT TRIGGER, maybe there are another Ideas?
regards

tggzzz:
If the inverter has a schmitt trigger input, it will generate a logic-level output.

If the inverter does not have a schmitt trigger input, then some types of inverters will be biassed into the linear region and act as a linear amplifier.

Choose your logic family, look at application notes. Try a circuit with the feedback resistor but without the capacitor, and see what happens. (And don't forget a decoupling capacitor!)

Zero999:

--- Quote from: yalect on October 13, 2019, 11:39:21 am ---thank you for your reply
but most not gate oscillator circuit works with two gate, one is SCHMITT TRIGGER, maybe there are another Ideas?
regards

--- End quote ---
Build it in real life to find out.

As mentioned above, the simulator is wrong. If it does oscillate in real life, it will be at a much higher frequency as many NOT gates, such as the 74HC04, are really three gates in series and will form a ring oscillator.


If you used the cd4069 which really is a single NOT gate, the output will simply sit at half the supply voltage and the supply current will be high, as both MOSFETs in the gate will conduct.

Navigation

[0] Message Index

There was an error while thanking
Thanking...
Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod