Electronically controllable negative capacitance (with NIC)

[nickagian]

Hi!

I would like to do the following:

I want a way to create a negative capacitance, whose value can be controlled electronically. One way that I have thought of doing this is to use a NIC with an input impedance of Zin = -kC, where C will be a known capacitor. If the k can be electronically controllable, then I have the controllable capacitance I want.

But I have some difficulties in how to make this k to be controllable.

(a)One solution would be to use k purely resistive (for example in an opamp NIC) but then I have the problem of how to create an electronically controllable resistance. Any idea here?

(b)Another solution would be to make factor k to depend on some biasing current/voltage. This maybe can be done with a transistor-based NIC? How can then the k depend on the biasing? Can anyone give me a schematic? I have searched in the internet, but haven't found something useful.

Can anyone help me with some ideas on the two points above? Or even a different idea on how to create the variable capacitance, negative or positive?

Thanks!

 

[FvM]

Without detail specification, one would think of the same circuit elements that can be used for tunable filters, analog multipliers, FETs as variable resistors, possibly LDRs. Frequency range, signal-to-noise ratio, linearity requirements, variation ratio and speed, accuracy are the parameters that decide about applicable techniques.

 

[nickagian]

FvM thanks for this information! Actually the current passive LC-network has a resonance frequency of around 1.5kHz and I would like to go lower than this frequency. Ideally down to 50 Hz but this is not necessary for the time being. Above that, I don't have any other specific requirement, like SNR or speed or accuracy. So I think there is a quite big flexibility regarding the range, since I can always put a fixed resistor in-series to the controllable and adjust the range as I want.

Actually I don't know anything about the techniques used in tunable filters. What about the FETs as variable resistors? Any FET will do the job or there exist specific component for this?

And what about the LDRs? Is there a technique to apply controllable light on it and control its resistance?

Thanks a lot!

 

[FvM]

Here's a previous edaboard thread discussing JFET as variable resistors. The technique can be basically applied to MOSFET as well.
https://www.edaboard.com/threads/129234/

LED+LDR combinations as analog optocouplers respectively variable resistors are available from several vendors, e.g. Perkin Elmer optolectronics or silonex Silonex Inc.: Products: Audiohm Optocouplers

 

[LvW]

Nickagian,

I like to try to complete the list of possibilities to find a resistor to be controlled electronically (as given by FvM in posting #2):

Another alternative is to use an OTA with 100% negative feedback. This circuit emulates a grounded resistor to be controlled with the stearing current Iabc.

 

[nickagian]

FvM,

thanks a lot for this information. For the time being this is enough to study and look for a possible implementation. If I have any more questions on the topic I will post again.

Nickagian,

I like to try to complete the list of possibilities to find a resistor to be controlled electronically (as given by FvM in posting #2):

Another alternative is to use an OTA with 100% negative feedback. This circuit emulates a grounded resistor to be controlled with the stearing current Iabc.

LvW,

Great! That sounds pretty much like what I wanted to use, but couldn't find how to implement it. Is it possible to give me some links where I can find more information about such an implementation? What is this current Iabc that you are talking about?

 

[LvW]

Hi Nickagian,

look here: ota-Tutorial1

(Fig. 3, first circuit with R=1/gm)

 

[nickagian]

Hi Nickagian,

look here: ota-Tutorial1

(Fig. 3, first circuit with R=1/gm)

Hi LvW,

thanks for this information. I have tried to implement both the grounded and the floating capacitance multipliers in LTSPice using the LM13700 OTA but with no success. Have you ever tried any of these? Do you know if I have to use the Darlington buffer or not/

Although the input impedance shows indeed capacitive behavior, the problem is that I cannot control its value at all. I keep the bias currents of all OTAs the same, except for one. I then change the bias current of this one, but the impedance does not change. This is not normal, right?

 

[LvW]

Nickagian,

my suggestion to use an OTA was related to the controllable (grounded) resistor only - to be used in an opamp based NIC circuit.
I did not speak about OTA-based NIC realizations in general.

 

[nickagian]

Nickagian,

my suggestion to use an OTA was related to the controllable (grounded) resistor only - to be used in an opamp based NIC circuit.
I did not speak about OTA-based NIC realizations in general.

I agree.. but nevertheless, have you ever used any of the other circuits, using OTA-based NIC or impedance multipliers in general? That was more or less my question. Of course it goes not only to you, but anyone who might read this post!

 

[LvW]

No sorry, I didn't. What about showing us your OTA-based NIC circuitry?