variable capacitor

[fifahaywire]

hi.
i've reviewed variable capacitors but the typical values for one exist with small range values (in pico farads etc). the range i need is about 0.007uF - 1uF. it doesn't matter if it varies more or less than this.
i've thought about picking out exact values and connect to which one i need. but i want to know if there is other solutions to this. anyone can help?

 

[unclejed613]

the largest ones i've ever seen made are about 500pf. what is this for? somewhere here i have a circuit to make a variable large value cap using transistors , a few resistors, a potentiometer and a cap. it can be ground referenced, or have both "ends" floating".

correction: found it, one end of the "capacitor" is grounded.

 

[ccurtis]

Sorry, you won't find mechanical variable capacitors that large. They would be physically very large and impractical.

The best I can think of is an op amp capacitance multiplier circuit in conjunction with available standard variable capacitors, but realize such circuits have the steep limitation that one end of the capacitor is connected to the power supply ground. They also have a steep limitation on how high a frequency they operate at. 10KHz and above gets iffy. They also exhibit larger series resistance and leakage current than physical capacitors.

The usual route is to use a switched capacitor bank similar to a resistor decade box. Depending on the application, you might be able to use electronic switches.

 

[unclejed613]

in the original text C1 was a 0.1uF cap. as such, the "capacitor at X and Y was variable between 0.1uF with the pot wiper at the ground end, and with it at the other end, the capacitance is that of the transistor junctions. +/-VCC can be anything from +/-7V to +/-35V (yes you need a bipolar supply)

 

[fifahaywire]

in the original text C1 was a 0.1uF cap. as such, the "capacitor at X and Y was variable between 0.1uF with the pot wiper at the ground end, and with it at the other end, the capacitance is that of the transistor junctions. +/-VCC can be anything from +/-7V to +/-35V (yes you need a bipolar supply)

hey! thanks for the reply. i'll see if i can try this out

 

[fifahaywire]

the largest ones i've ever seen made are about 500pf. what is this for? somewhere here i have a circuit to make a variable large value cap using transistors , a few resistors, a potentiometer and a cap. it can be ground referenced, or have both "ends" floating".

correction: found it, one end of the "capacitor" is grounded.

basically this is for an oscillator. i'm trying to vary the capacitor values to get different frequencies. about 5k - 20kHz
can the circuit you give be applied?

Sorry, you won't find mechanical variable capacitors that large. They would be physically very large and impractical.

The best I can think of is an op amp capacitance multiplier circuit in conjunction with available standard variable capacitors, but realize such circuits have the steep limitation that one end of the capacitor is connected to the power supply ground. They also have a steep limitation on how high a frequency they operate at. 10KHz and above gets iffy. They also exhibit larger series resistance and leakage current than physical capacitors.

The usual route is to use a switched capacitor bank similar to a resistor decade box. Depending on the application, you might be able to use electronic switches.

yeah, i was thinking about the switching capacitor bank too. it seems to be the easiest way. but the values available will be limited. that's why i'm trying to look for other ways. i'll definitely look into the op amp capacitance multiplier circuit. thanks!

 

[unclejed613]

the transistor version should work well beyond 20khz. also, if this is for a sine wave generator, be aware that just taking a quick look at the way the transistors are connected together, suggests to me there may be a "dead zone" near the zero crossing. in an amplifier i would call it a crossover notch. if you can live with a little bit of distortion, it should be ok. there are op amp circuiys where both ends of the "capacitor" are floating. this transistor circuit dates to about 1968. op amps were not yet available on single chips.

 

[audioguru]

I made a Wien bridge oscillator. It goes from about 10Hz to 100kHz in four ranges from 10Hz to 100Hz, 100Hz to 1kHz, 1kHz to 10kHz and 10kHz to 100kHz. It uses a dual 35pF to 365pF tuning capacitor from an old AM radio.
Since the capacitance is low then the resistor values are very high so it uses a Jfet-input opamp.

Since you want to go from 5kHz to 20kHz then a Wien bridge oscillator using the dual variable capacitor from an old AM radio would work fine with a Jfet-input opamp.

Post your schematic for us to see what kind of oscillator you have.