Modulating frequency of Multi-vibrator circuit (CMOS technology)

Thread Starter

Alex_Khan

Joined May 27, 2020
60
Hello,
I am modulating control signal with Multivibrator circuit in CMOS technology. I use this formula (1/2.2RC) to find the frequency of modulated signal with multivibrator. But my simulation (result) and theoritical (calculation) of frequency is not matching. I am using CADENCE software tool for my simulation and have attached the circuit diagram along with simulation result for your refrence.

Calculation:
R=287k ohm.
C=27pF
Freq(1/2.2RC)=58.6KHz

Simulation:
Freq: 55.9kHz

Questions:
Am i using the right formula/Simulations for calculation?
Why is the difference between calculation and simmulation results?
How to get exactely matched both the results?

Please someone can sort out this problem.

Thanks.
 

Attachments

BobTPH

Joined Jun 5, 2013
8,765
I am surprised it is that close, actually. How to get it to natch? Change the formula so that it matches. You will likely get results even further off when built with real components, because they never match the simulation exactly.

Bob
 

DickCappels

Joined Aug 21, 2008
10,140
Also keep in mind that the simulator is not using a formula to know what the frequency is, it is simulating the behavior of the components.

As @BobTPH suggested just adjust the constant (2.2) in your formula to get a match
With the simulator.

Having had a circuit like this in high volume production I expect that you will see much greater variation from part to part. The production processes for most CMOS gates are not really well controled for analog parameters.

If you put a large resistor in series with the feedback to the NAND gate that would improve temperature dependence and quite possibly make the simulation more satisfying because that would reduce the effects of input protection diodes occasionaly going into conduction.
 

Thread Starter

Alex_Khan

Joined May 27, 2020
60
I am surprised it is that close, actually. How to get it to natch? Change the formula so that it matches. You will likely get results even further off when built with real components, because they never match the simulation exactly.

Bob
Thanks for your response ! your explaination helps me to understand the situation.
 

Thread Starter

Alex_Khan

Joined May 27, 2020
60
Also keep in mind that the simulator is not using a formula to know what the frequency is, it is simulating the behavior of the components.

As @BobTPH suggested just adjust the constant (2.2) in your formula to get a match
With the simulator.

Having had a circuit like this in high volume production I expect that you will see much greater variation from part to part. The production processes for most CMOS gates are not really well controled for analog parameters.

If you put a large resistor in series with the feedback to the NAND gate that would improve temperature dependence and quite possibly make the simulation more satisfying because that would reduce the effects of input protection diodes occasionaly going into conduction.
Thanks for your explaination. i will try the resistor and then analyze its result.
 
Top