CCCS circuit for ADS simulation

[erikl]

I simulated the Mosfet circuit with try and error method and draw real part of input impedance vs frequency but I did not get the desired response yet (it isnt negarive and is positive for all frequency)

It seems this circuit doesn't work as a NIC, probably because the additional currents needed to create the correct reference points (i.e. due to the multiple current mirroring). I must admit I didn't know in the very beginning of this thread that you need a NIC circuit.

I tried the circuit with supposed technology currents and Kp values for your process and a DC input current of 10mA (it works between 1..100mA).
To get reasonable operation points, I had to scale up your resistor values by a factor of 1000 (and still had to use ridiculous large MOSFET widths), s. this PDF: View attachment CCCS3.pdf
With the given DC input current of 10mA, the pos. input impedance is 7.7Ω .

 

[electronicman26]

It seems this circuit doesn't work as a NIC, probably because the additional currents needed to create the correct reference points (i.e. due to the multiple current mirroring). I must admit I didn't know in the very beginning of this thread that you need a NIC circuit.

I tried the circuit with supposed technology currents and Kp values for your process and a DC input current of 10mA (it works between 1..100mA).
To get reasonable operation points, I had to scale up your resistor values by a factor of 1000 (and still had to use ridiculous large MOSFET widths), s. this PDF: View attachment 86422
With the given DC input current of 10mA, the pos. input impedance is 7.7Ω .

Thank you for the time you spent helping me
now dont you have any idea for replacing ideal CCCS at first circuit and it remain and acts as a NIC?

 

[erikl]

Thank you for the time you spent helping me
now dont you have any idea for replacing ideal CCCS at first circuit and it remain and acts as a NIC?

Unfortunately not. Try and search G00GLE for real-device NICs!
Good luck!

 

[electronicman26]

Unfortunately not. Try and search G00GLE for real-device NICs!
Good luck!

OK, Thanks
but my problem isnt NIC , I have problem with circuitary model that replace to ideal CCCS

 

[erikl]

... my problem isnt NIC , I have problem with circuitary model that replace to ideal CCCS

Just here this part of my circuit is a good example:

The current injected into T3 (≈2µA) is repeated via T4, T10, T11, T12 (all the same current, but now referred to a different node), then translated via T13 with the required transmission ratio of 1.5 ((W13/L13) / (W12/L12)), now ≈3µA to the load resistor.

Isn't this what you wanted?

 

[electronicman26]

Just here this part of my circuit is a good example: View attachment 86479

The current injected into T3 (≈2µA) is repeated via T4, T10, T11, T12 (all the same current, but now referred to a different node), then translated via T13 with the required transmission ratio of 1.5 ((W13/L13) / (W12/L12)), now ≈3µA to the load resistor.

Isn't this what you wanted?

Hello erikl
as clear in below schematic:

the current to the load resistor is (A'+1)i2 that is (1.5+1)*2um, and it seem good in your circuit, maybe the problem is on the first CCCS configuration beacuse its clear that the above circuit should act as a NIC and the simulation with ideal CCCS prove it

 

[erikl]

... beacuse its clear that the above circuit should act as a NIC and the simulation with ideal CCCS prove it

Yes, and it's also clear that my MOSFET circuit will not act as a NIC, because it has to transfer the initial currents (from T1 & T3, s. my CCCS3.pdf from above) to different reference nodes (which is no problem with CCCS), and by this creates additional currents which contribute to i₁.

 

[electronicman26]

Yes, and it's also clear that my MOSFET circuit will not act as a NIC, because it has to transfer the initial currents (from T1 & T3, s. my CCCS3.pdf from above) to different reference nodes (which is no problem with CCCS), and by this creates additional currents which contribute to i₁.

ok dear erikl, where is the problem you think? and now how I can replace ideal CCCS with circuitary models that the final two-port act as a NIC? do you have any idea for this implementation?

 

[erikl]

where is the problem you think?

You would know if you had read my above explanation thoroughly.

how I can replace ideal CCCS with circuitary models that the final two-port act as a NIC? do you have any idea for this implementation?

Sorry, no, this isn't my concern, it's your business. I already told you above where to go on searching. Good bye and good luck!

 

[electronicman26]

You would know if you had read my above explanation thoroughly.

OK, but your proposed CCCS dont work like Ai1 with 2 port, if all of circuitary model is like the ideal form then why the operation of them are different

Sorry, no, this isn't my concern, it's your business. I already told you above where to go on searching. Good bye and good luck!

yes, I know, but at first I searched in google and beacuse of I could not find anything I post my question here