Hi,

Currently I need a voltage controlled capacitor that has value between 1nF to 2μF. Besides connecting the varactor (which has low capacitance value ~ pF range) in parallel, anyone has any idea on how to achieve the mentioned values using a different approach? Appreciate if you can share useful information. Thanks

Regards,
lkgan

 

My guess would be a switching setup where you can switch in several decades of fixed capacitors. You might need to use mechanical switches. Perhaps a couple of thumbwheel switches to select the desired capacitance.

This would work for a lab capacitance box, but not for much else.

 

Well, I actually plan to control the capacitance value using a microcontroller for automation purposes, hence need to use voltage to control it. Any suggestions for it?

 

What flashed into my head was a gigundus gang variable capacitor with a servo turning it. Not very practical.

 

Could you use ganged caps switched in and out with a 4066 analog switch?

 

Use caps connected to a microcontroller pin. Make the pin hi-Z (tristated) to remove the capacitance; set the output low to use the capacitance.

 

Hello,

What voltage is expected on the "variable" capacitor?
What frequency is the voltage accross the "variable" capacitor?

Bertus

 

What sort of Q or loss angle do you need, and over what frequency range? If you are going to use the approach of switching capacitors using analogue switches, micro-controller pins etc, the switch "on" resistance may have a big effect.

For instance, the CD4066 might have 80 ohms on resistance, but the reactance of 1μF at 1kHz is only 159 ohms - not very high Q!

In addition, the current carried by the capacitors must be within the ratings of any switches you use.

 

You can also research on opamp based "Capacitance multiplier".

An image from Wiki:

 

Hi,

Currently I need a voltage controlled capacitor that has value between 1nF to 2μF. Besides connecting the varactor (which has low capacitance value ~ pF range) in parallel, anyone has any idea on how to achieve the mentioned values using a different approach? Appreciate if you can share useful information. Thanks

Regards,
lkgan

What are you really trying to do, i.e., why do you need a voltage controlled capacitance?
We frequently get questions wherein the OP has already decided on a general solution to a problem but needs a piece of "unobtanium" to complete the design. There may be another approach to solving your problem that does not require a part that you cannot get.

Having said that, there are probably ways to make voltage-controlled high capacitances, but they probably involve analog multipliers, which are expensive and require a high degree of circuit design capability.

 

Thanks everyone for giving possible solutions. I am actually trying to vary the variable capacitor for filtering purposes. Here comes my answer for others.

Could you use ganged caps switched in and out with a 4066 analog switch?

It is possible to use analog switch to choose different capacitors, but I am afraid that the switching might introduce noise. This might also adds more complexity to the whole design to be bulky.

Use caps connected to a microcontroller pin. Make the pin hi-Z (tristated) to remove the capacitance; set the output low to use the capacitance.

Can you please explain more on this concept (probably with a diagram for better illustration), don't really get the idea.

Hello,

What voltage is expected on the "variable" capacitor?
What frequency is the voltage accross the "variable" capacitor?

Bertus

The expected voltage should be less than 16V.
The frequency is in kHz to MHz range.

What sort of Q or loss angle do you need, and over what frequency range? If you are going to use the approach of switching capacitors using analogue switches, micro-controller pins etc, the switch "on" resistance may have a big effect.

For instance, the CD4066 might have 80 ohms on resistance, but the reactance of 1μF at 1kHz is only 159 ohms - not very high Q!

In addition, the current carried by the capacitors must be within the ratings of any switches you use.

Thanks for mentioning the possible problems for using analog switches. Generally the Q doesn't really matter for now, since I am not sure how the Q will affect the overall performance.

You can also research on opamp based "Capacitance multiplier".

An image from Wiki:

Thanks for giving a new solution that I have never thought of. I think the op-amp introduces extra noise to my design, might not be able to use it and it's quite expensive to have this design.

What are you really trying to do, i.e., why do you need a voltage controlled capacitance?
We frequently get questions wherein the OP has already decided on a general solution to a problem but needs a piece of "unobtanium" to complete the design. There may be another approach to solving your problem that does not require a part that you cannot get.

Having said that, there are probably ways to make voltage-controlled high capacitances, but they probably involve analog multipliers, which are expensive and require a high degree of circuit design capability.

Is the analog multiplier you mentioned same as the one suggested by eblc1388 using op-amp approach?

 

Hello,

Can you post a schematic of the intended circuit?

16 Volts is to high for analog switched.
Is the capacitor grounded?
If so, than you could probably use mosfets for the switching.

Bertus

 

Hi Bertus,

Sorry to say that I actually haven't design the schematic, it's just a concept in my mind, I am still open for ideas. It's for filtering purposes. MOSFET switching approach is a choice but I will need to check whether it will contribute extra noise to my system.

 

Is the analog multiplier you mentioned same as the one suggested by eblc1388 using op-amp approach?

No. Eblc1388's circuit does not have voltage control capabilities. Perhaps he was think that you could replace the capacitor with a varicap, but it would be a big stretch to get from the capacitance of a varicap to the range you want.
Also, varicaps are two-terminal devices, so their capacitance is modulated by the signal. This is not a good thing if you are working with high-level signals.

 

No. Eblc1388's circuit does not have voltage control capabilities. Perhaps he was think that you could replace the capacitor with a varicap, but it would be a big stretch to get from the capacitance of a varicap to the range you want.

Well, apologize that I did not justify clearly before this. The range I mentioned can be split into few particular small ranges. Perhaps 0.1nF to 10nF for a varicap. And another one would be 1μF to 2μF, something like that. Of course it will be ridiculous for me to design a varicap that works 1nF to 2μF

 

What order filter do you want? AFAIK, it is easier to make a voltage controlled resistor than it is a voltage controlled capacitor.
Another possibility is switched-capacitor filters, but I don't think you can get them that will work at MHz signal frequencies.

 

No. Eblc1388's circuit does not have voltage control capabilities. Perhaps he was think that you could replace the capacitor with a varicap, but it would be a big stretch to get from the capacitance of a varicap to the range you want.

Actually no.

I was thinking about the possibility of using large value capacitor and changing digital pot taps to get what the OP wants.

 

Lkgan, If you can be happy with a tunable filter that will not work past 100kHz, have a look at LMF100. You probably won't find a simpler way to tune a filter. Tuning is done by varying the clock frequency.

 

Take a look at GYRATOR CIRCUITS , ask google about it.


good luck
Lineman

 

Take a look at GYRATOR CIRCUITS , ask google about it.


good luck
Lineman

The key is voltage controllability (is that a word?).